University of Birmingham Research Archive
e-theses repository This unpublished thesis/dissertation is copyright of the author and/or third parties. The intellectual property rights of the author or third parties in respect of this work are as defined by The Copyright Designs and Patents Act 1988 or as modified by any successor legislation. Any use made of information contained in this thesis/dissertation must be in accordance with that legislation and must be properly acknowledged. Further distribution or reproduction in any format is prohibited without the permission of the copyright holder.
5th of 5 files
Chapters 7 and 8 Appendix and References
EFFECT OF STRESS ON INITIATION AND PROPAGATION OF LOCALIZED CORROSION IN
ALUMINIUM ALLOYS By
SUKANTA GHOSH
A thesis submitted to University of Birmingham
for the degree of DOCTOR OF PHILOSOPHY
Metallurgy and Materials School of Engineering
University of Birmingham
November 2007
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
346
7 X-RAY SYNCHROTRON TOMOGRAPHIC STUDY OF
LOCALIZED CORROSION PROPAGATION IN
ALUMINIUM ALLOYS
X-ray synchrotron tomography is a relatively new non-destructive technique
which has found ever increasing use in various fields of materials science and engineering
for acquiring three dimensional (3D) information of materials. New generation
synchrotron sources are able to deliver reconstruction of three dimensional images with
spatial resolution close to 1 µm. Because of the resolution in the micron or sub micron
level, X-ray synchrotron tomography is sometime referred as X-ray synchrotron
microtomography. Basic theory of x-ray synchrotron tomography and its advantages over
conventional two-dimensional techniques has been discussed in Section 2.4 (Chapter 2)
and can also be found elsewhere [261-267]. Details of the experimental set up for
performing X-ray synchrotron experiments can be found in Chapter 3. X-ray
microtomography can overcome the limitations of 2D techniques (e.g. true size
distribution and connectivity of phases with complex shapes) and provides information
about what happens in the bulk material [268]. Reconstruction of the 3D volumetric data
from this technique allows for the analysis of the microstructure, defects, phase
distribution or damage in the bulk material along with the ability to view two dimensional
cross sections at any particular plane or orientation in a non-destructive manner.
The key purpose of this present study is to examine the effect of stress on growth
and development of localized corrosion (i.e., intergranular corrosion) in a high strength
aluminium alloy. Aerospace aluminium alloy AA2024 in a sensitised temper condition
was used for this study. AA2024 specimens were heat treated for 2 hours at 250˚C
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
347
followed by water quenching. This particular heat treatment had shown to simulate the
electrochemical behaviour of the heat affected zone (HAZ) in the friction stir welded
plate [70]. Details of the heat treatment can be found in Chapter 3 and further synchrotron
experiments with this particular sensitized temper in unstressed condition is reported
elsewhere [235, 288]. Yield strength (Y.S.) of the heat treated AA2024 was determined to
be approximately 375 MPa. Experiments were performed on both unstressed and stressed
(70% Y.S. and 90% Y.S.) samples.
The evolution of intergranular corrosion (IGC) in aluminium alloys is a problem
of fundamental interest to the corrosion community for several years and also somewhat
poorly understood. So far only 2D techniques have been used for growth rate
quantifications. However, because of the limitations of the conventional two dimensional
techniques it is difficult to measure a 3D phenomena such as IGC evolution inside the
bulk alloys. So, in this study, high resolution in situ X-ray synchrotron microtomography
has been used to record the three dimensional evolution of localized corrosion sites in an
aqueous chloride solution, as a function of applied stress and exposure time. Outcome of
this work will provide quantitative growth rates of intergranular corrosion and thereby
significantly contribute to the ongoing modelling effort for life prediction of aircraft
components.
7.1 Two Dimensional Visualization of Localized Corrosion Morphology
Figure 7.1a and b shows the examples of two reconstructed X-ray tomographic
slices from a single plane of an unstressed and stressed (90% Y.S.) bulk AA2024
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
348
sensitized specimens after exposure of 15 h and 22.4 h, respectively. These reconstructed
images (i.e., Figure 7.1a-b) are obtained from the in situ exposure in 0.6 M NaCl solution
during the synchrotron experiments. Several features of the bulk alloy can be revealed
from these tomographs30 without actually destroying the specimen. As these tomographs
are generated using absorption mode31, different atomic number of different elements
inside the bulk alloy creates contrast in the tomographs helping to provide good
visualization of intermetallic particles and intergranular corrosion sites. It can be seen
from Figure 7.1(a) that the higher atomic number intermetallic particles appear bright in
the microstructure. These intermetallic particles sometimes appear as a band aligning
parallel to the rolling direction.
As the specimens are made from a rolled plate, three different directions are
assigned to each of the samples, longitudinal (L), transverse (T) and short transverse
(S).32 Figure 7.2 shows the schematic of a typical specimen with these three different
orientations. It can be seen that the longer side of the specimen is in the transverse (T)
direction. Stressed samples were subjected to load in the transverse (T) direction. Figure
7.2 also schematically explains how a single slice from a single plane is represented
within the bulk specimen. Figure 7.1(a) and (b) show that the attacks are intergranular
(IGC) in nature and normally follow the L direction which is also the rolling direction.
Pre-existing cavities (predominantly associated with the intermetallic particles) within the
bulk alloy can also be easily identified from these tomographs.
30 A tomograph is a 2D slice or plane within the 3D reconstructed X-ray attenuation data. 31 Different modes of X-ray synchrotron tomography is described in Chapter 3. 32 Procedure of making ‘matchstick’ cylindrical specimen and dog-bone specimen from the rolled AA2024
plate is described in Chapter 3.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
349
Figure 7.1 Reconstructed tomographic slice of a single plane through the sensitized AA2024
specimen (i.e., heat treated at 250˚C for 2h) after in situ open circuit exposure in 0.6 M
NaCl, (a) Unstressed, after 15 h exposure and (b) Samples stressed to 90% YS, after 22.4 h
exposure. L = Longitudinal Direction, S = Short Transverse Direction.
L
S
L
S
(a)
(b)
Intermetallic Particles
The Alloy appears to contain cavities, particularly adjacent to the intermetallic particles
Intergranular corrosion
Corrosion attacks follow the rolling direction
Some of the surface remains unattacked
Machined outer surface is unattacked whereas metal underneath the deformed layer is dissolved
Hydrogen bubble in the electrolyte generated during corrosion reaction
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
350
Figure 7.2 Schematic of a typical experimental specimen of sensitized AA2024 sample with
different orientations. L = Longitudinal Direction, S = Short Transverse Direction, and T =
Transverse Direction. Stress is applied in the transverse (T) direction. It should be noted
that for unstressed condition cylindrical matchstick specimen and under stressing condition
dog-bone shaped specimens were used. However, in both cases specimen orientations are
represented in the way represented in the schematic.
It is interesting to note that all intermetallic particles are not attacked and results in
IGC (see Figure 7.1a). This is evident from the fact that in many cases intermetallic
particle remain unattacked on the alloy surface. Hydrogen bubbles can also be seen in the
solution surrounding the specimen. These hydrogen bubbles are generated from the
supporting cathodic reaction during the corrosion process.
Figure 7.1b shows extensive IGC attacks of the sensitized AA2024 specimen
stressed to 90% Y.S after 22.4 h of exposure. During the later stages of the localized
corrosion event, the evolution of the IGC is associated with considerable amount of grain
interior dissolution and/or fallout of grains. One interesting observation in this
tomographic slice is that, at some places of the sample surface, the corrosion rate seems
T
L
S
T
L
S
T
L
S
Reconstructed single tomographic slice
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
351
lower than that of the bulk of the material. These differential corrosion rates in two
adjacent areas result in considerable undercutting of the specimen. This may be an
indication that the machining process for manufacturing the specimens can actually affect
the corrosion susceptibility and the rate of local attacks on the surface [288].
Progressive development and coalescence of localized corrosion sites within the
bulk of unstressed and stressed AA2024 specimens (in a sensitized temper condition)
have been studied through in situ X-ray tomographic scans of individual specimens at
four different times during continuous exposure in naturally aerated 0.6 M NaCl solution.
Tomographic scans as a function of exposure time not only help to quantify the rate and
mode of intergranular attack, but it also gives the opportunity to visualize any possible
interaction between the constituent particles and propagating localized corrosion attacks.
Tomographic images of the interaction between a growing IGC site with an
intermetallic particle within the bulk of an unstressed AA2024 specimen as a function of
exposure time is shown in Figure 7.3. It is interesting to note that IGC initiated adjacent
to an intermetallic near the outer surface of the specimen. However, it should also be
remembered that IGC does not always initiate near all of the intermetallic particles
exposed to the sample surface. These observations re-emphasize the fact that localized
corrosion initiation in a material is a stochastic process [18]. Large (size range 10-30 µm)
and irregular shaped intermetallic particles in AA2024 are identified as Fe-Mn particles
which act as cathodic sites (i.e., provide better surface for cathodic reaction) during the
corrosion process.33 Hence some intermetallic particles remain unattacked even after 34
h of exposure in 0.6 M NaCl (Figure 7.3d). Intergranular corrosion around the
intermetallic particle grows without affecting the particle. Dissolution of the surrounding 33 Morphology and corrosion properties of different intermetallic particles present in AA2024 is described
in details in Chapter 4 and 5.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
352
alloy matrix can easily be seen. IGC sites propagate in the L direction and simultaneous
widening of the localized corrosion site occurs. Gradual widening of two adjacent
intergranular corrosion sites due to grain dissolution result into coalescence of corrosion
sites. Hydrogen bubbles adjacent to the corroding alloy [Figure 7.3b, c and d] results from
the supporting cathodic reaction.
Figure 7.3 A series of reconstructed tomographic slices of a particular plane through the
sensitized AA2024 specimen (i.e., heat treated at 250˚C for 2h) in unstressed condition after
in situ open circuit exposure in 0.6 M NaCl for (a) 3.4 h , (b) 15 h, (c) 25.6 h, and (d) 33.9 h.
These tomographs illustrate the interaction between a propagating intergranular attack and
intermetallic particles.
Localized corrosion evolution at a particular site in a single plane of tomographic
slices of an unstressed cylindrical AA2024 sample (sensitised tempered) is shown in
Attack around the intermetallic particle Dissolution of the alloy around the intermetallic particle
(a) (b)
(c) (d)
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
353
Figure 7.4. Series of these exposures prove that localized corrosion does not initiate in or
adjacent to all exposed constituent particles. Even after 34 h of exposure in 0.6 M NaCl
solution new corrosion sites did not develop. IGC initiated only at a single site and
propagated as can be seen from the tomographic images (Figure 7.4). Normally it is found
that initially IGC attack follows the longer grain paths in the rolling direction (L). IGC
growth in L direction and broadening of those localized corrosion sites occur
simultaneously during the exposure. However, intergranular attack stops growing in the L
direction after 25 h of immersion (Figure 7.4c) and starts widening at a faster rate which
may be caused by grain dissolution and/or grain fall out. Hydrogen bubbles associated
with the corrosion of aluminium can be seen at the site of localized corrosion. Fox [235]
performed both ex situ study of simulated HAZ and in situ study of a HAZ of a friction
stir welded AA2024-T3 in aerated 0.6M NaCl under open circuit condition. Results from
Fox’s study also emphasized that localized corrosion does not necessarily initiate at all
constitute particle and after a certain amount of growth, IGC stops growing (or grows at a
much reduced rate) in the L direction but continues to corrode locally via subsequent and
considerable grain loss/or fall out adjacent to the IGC sites.
Figure 7.5 shows 2D tomographs of reconstructed X-ray attenuation data of IGC
evolution at a particular site in a single plane in the sensitized tempered AA2024 sample
stressed to 70% of its yield strength (YS). Two predominant sites are found to be
evolving as a function of time, one in the top left corner and second one in the bottom of
the images shown in Figure 7.5. Growth of the top left IGC attack is found to cease (or
significantly decrease) in the L direction after 16.9 h. This observation is quite similar to
that found in unstressed sample.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
354
Figure 7.4 Reconstructed 2D tomographic images showing the evolution of IGC at a
particular site in the sensitized AA2024 specimen (i.e., heat treated at 250˚C for 2h) in
unstressed condition after in situ open circuit exposure in 0.6 M NaCl as a function of time,
i.e., (a) 3.4 h, (b) 15 h, (c) 25.6 h, and (d) 33.9 h of exposure. [Hydrogen bubble resulting
from the supporting cathodic reaction on the specimen can be seen in each tomograph]. L =
Longitudinal direction which is also rolling direction, S = Short Transverse Direction.
(a) (b)
(c) (d)
L
S
Intermetallic Particles
Intergranular Corrosion
(a) (b)
(c) (d)
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
355
Figure 7.5 Reconstructed 2D tomographic images showing the evolution of IGC at a
particular site in the sensitized (i.e., heat treated at 250˚C for 2h) and stressed (70% yield
strength) AA2024 specimen after in situ open circuit exposure in 0.6 M NaCl as a function
of time, i.e., (a) 5.6 h, (b) 16.9 h, (c) 27.6 h, and (d) 35.9 h of exposure. IGC stops growing in
‘L’ direction after certain time and new IGC sites initiates (c). L = Longitudinal direction
which is also rolling direction, S = Short Transverse Direction. Stress is applied in
transverse direction, T.
(a) (b)
(c) (d)
Intergranular Corrosion
IGC stops growing in ‘L’ direction Initiation of new IGC sites
L
S
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
356
The IGC network at the bottom of the tomographic slices has virtually penetrated
through bulk of the sample and reached the other exterior surface of the specimen within
the first six hours of exposure (Figure 7.5c). Another corrosion site (which could be new
or may just be growing in the T direction in 3D) appears after 27.6 h in the middle right
side of Figure 7.5c. As seen in unstressed samples, IGC in stressed sample also stops
growing in L direction and broadening of IGC sites occur through grain interior
dissolution and/or grain fall out.
Volume analysis in three dimensions (3D) showed an increase in volume loss as a
function of time which proves that, though IGC stop growing (or significantly decreases
growth rate) in the L direction it continues to grow in other directions. It may be possible
that in both stressed and unstressed samples in many cases localized corrosion initiates as
pitting attacks which later develop as intergranular corrosion.
Figure 7.6 shows the tomographs at a particular site in a single plane of the
AA2024 sample stressed to 90% of its yield strength as a function of exposure time in 0.6
M NaCl. A significant amount of attacks have been observed after 10.9 h of exposure
(Figure 7.6a) and after 22.4 h of exposure propagating IGC sites start widening via
dissolution of the grain interior adjacent to the IGC sites. It is interesting to note that
(Figure 7.6c), though the outer surface of the sample is still intact in some portions,
significant IGC occurred underneath the surface. This undercutting suggests that there are
some areas on the surface of the specimen where the corrosion rate is lower than the bulk
of the material. One possible explanation for this undercutting is that the machining
operation for manufacturing those specimens possibly has an effect on the corrosion
susceptibility and local rate of attack on the surface [288].
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
357
Effect of applied stress on the intergranular corrosion of heat treated AA2024
specimens can be visualized from three representative tomographs of three different
samples shown in Figure 7.7. All samples were exposed in 0.6 M NaCl under open circuit
corrosion conditions. The tomograph of the unstressed sample (Figure 7.7a) represent
corrosion at a particular plane within the specimen after 15 hours of exposure, where as
70% Y.S. (Figure 7.7b) and 90% Y.S. (Figure 7.7c) tomographs are after 17 h and 11 h of
exposure respectively. It should be noted that though tomographic images of three
different samples with three different exposure times are compared in this figure, the
effect of stress is quite prominent. More IGC attacks can be seen in the sample stressed to
90% Y.S. even with shorter exposure time than the unstressed or 70% Y.S. specimens.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
358
Figure 7.6 Reconstructed 2D tomographic images showing the evolution of IGC at a
particular site in the sensitized (i.e., heat treated at 250˚C for 2h) and stressed (90% yield
strength) AA2024 specimen after in situ open circuit exposure in 0.6 M NaCl as a function
of time, i.e., (a) 10.9 h, (b) 22.4 h, (c) 33.2 h, and (d) 41.2 h of exposure. Severe corrosion
attacks can be seen after 22 h of immersion, though some portions of the surface remain
unattacked. L = Longitudinal direction which is also rolling direction, S = Short Transverse
Direction. Stress is applied in transverse direction, T.
L
S
(a) (b)
(c) (d)
Intergranular Corrosion
Undercut of machined surface layer Some part of the surface remain unattacked
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
359
Figure 7.7 Reconstructed 2D tomographic images showing the effect of stress on the
intergranular corrosion evolution in sensitized (i.e., heat treated at 250˚C for 2h) AA2024
specimens after in situ open circuit exposure in 0.6 M NaCl, (a) Unstressed sample after 15 h
of exposure, (b) Sample stressed to 70% Y.S. after 16.9 h of exposure, and (c) Sample
stressed to 90% Y.S. after 10.9 h of exposure. It should be noted that though tomographic
images of three different samples with different exposure time are compared in this figure,
effect of stress is prominent. More IGC attacks have been seen in 90% YS sample even with
shorter exposure time than the other two specimens. L = Longitudinal direction which is
also rolling direction, S = Short Transverse Direction. Stress is applied in transverse
direction, T.
(a)
(b)
(c)
Without Stress
70% Y.S.
90% Y.S.
L
S
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
360
7.2 Three Dimensional Visualization of Localized Corrosion Morphology
Two dimensional tomographs can be combined together in a stack to provide a 3D
representation of the IGC morphology and its interaction with the alloy microstructure.34
Figure 7.8 shows such 3D rendering of the tomographic data for ~ 150 slices of corroded
areas from different samples. Top and side views of these 3D images confirm the
intergranular nature of the attack without any presence of cracking. Specimen matrix
material has been rendered translucent in order to observe the intergranular attacks which
are labelled in green within the interior of the specimen.
It is clearly visible from Figure 7.8 that localized corrosion attacks are following
the grain boundaries of the elongated grains in the rolling direction. Several localized
corrosion sites initiate along the exposed section of the specimen (T direction) and grow
in the L direction in parallel (refer back to Figure 7.2 for orientation details in a
specimen). Several of these localized corrosion sites coalesce and form a plate like IGC
network in the transverse direction as can be seen in the side view from Figure 7.8.
Prolonged exposure of the samples in the aggressive solution will broaden the IGC
attacks which will result in grain interior dissolution and/or fall out. The nature and
morphology of intergranular attacks in this current study with and without application of
stress are very similar to that found by Fox [235] in unstressed sample under similar
experimental conditions. These similarities emphasized the fact that stress is actually
increasing the intergranular corrosion kinetics rather than promoting cracking as such.
34 Details of the 3D volume generation from tomographic slices and the software used for this process are
described in Chapter 3.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
361
50 µm 50 µm
75 µm 75 µm
200 µm 200 µm
(a)
(b)
(c)
Top View Side View
50 µm50 µm 50 µm50 µm
75 µm75 µm 75 µm75 µm
200 µm200 µm 200 µm200 µm
(a)
(b)
(c)
Top View Side View
Figure 7.8 Three dimensional reconstruction of some attacked areas (about 150
tomographic slices in each case) in sensitized AA2024 specimens showing top and side view
of the intergranular nature of the attacks during their exposure in 0.6M NaCl. (a) Sample
with no-stress after 15 h of exposure, (b) 70% YS sample after 16.9 h of exposure, and (c)
Sample with 90% YS after 10.9 h of exposure. IGC in 150 slices has been labelled in green.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
362
Figure 7.9 shows the SEM images of the unstressed sensitized temper (i.e., heat
treated at 250˚C for 2h) AA2024 specimen after open circuit exposure in 0.6 M NaCl for
35h.35 The intergranular nature of the corrosion can clearly be seen from these SEM
images along with significant grain dissolution and/or grain fall out. Figure 7.9a shows a
large attacked area due to extensive grain dissolution/fall out on the cylindrical sample
surface. Closer examination on some of the attacked area reveals the presence of more
distinct intergranular corrosion (Figure 7.9b-f). Figure 7.9c shows the grain facets in a
corroding area after some grains have fallen out due to corrosion. The lateral spread of the
intergranular corrosion is speculated to be higher than the depth of the intergranular
corrosion attack. These observations are consistent with the fact that corrosion penetration
might cease or be limited along the ‘L’ direction after sometime and different corrosion
sites initiated along the ‘T’ direction on the sample surface coalesce.
Figure 7.10, Figure 7.11, and Figure 7.12 show the 3D reconstruction of the
unstressed, 70% Y.S. and 90% Y.S. specimens respectively as a function of exposure
time in 0.6 M NaCl. The specimen matrix material has been rendered translucent in order
to observe the growth of localized corrosion sites which are labelled in pink colour within
the interior of the specimen. Top and side views of the reconstructed samples have been
shown to provide a better idea about the evolution of corrosion morphology as well as the
location of corrosion attacks.
35 SEM analysis was performed on the unstressed sample after its last exposure in synchrotron. Sample was
taken out from the environmental cell after the completion of synchrotron experiment and cleaned in
ethanol. Due to some time lag between the completions of synchrotron experiments and SEM analysis, the
sample under SEM may show slightly more corrosion attacks than found from the last scan in synchrotron.
However, due to significant amount of corrosion, stressed specimens had lost their mechanical integrity and
disintegrated while removing from the stress-rig. So, SEM analysis could not be performed on the stressed
samples.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
363
Figure 7.9 SEM images of the unstressed and sensitized (i.e., heat treated at 250˚C for 2h)
AA2024 specimen after in situ open circuit exposure in 0.6 M NaCl for 35 h. SEM images
are taken after the synchrotron tests. Intergranular nature of the corrosion attacks can
clearly be seen from the micrographs. Extensive intergranular corrosion leads to grains fall
out from the sample surface and turn into deep pits or holes on the surface at the later stage
of exposure. Crystallographic facets of different grains can also be seen at the base of most
of the attacks, (c) and (f).
(a) (b)
(c) (d)
(e) (f)
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
364
Figure 7.10 Top and side views of the three dimensional reconstructed images (about 1000
tomographic slices in each case) showing the evolution of IGC in the sensitized AA2024
specimen (i.e., heat treated at 250˚C for 2h) in unstressed condition after in situ open circuit
exposure in 0.6 M NaCl as a function of time, i.e., (a) 3.4 h, (b) 15 h, (c) 25.6 h, and (d) 33.9 h
of exposure. Intergranular corrosion has been labelled in pink colour.
Side View Top View (a)
3.4 h
(b)
15 h
(c)
25.6 h
(d)
33.9 h
300 µm
TL
S
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
365
Figure 7.11 Top and side views of the three dimensional reconstructed images (about 1000
tomographic slices in each case) showing the evolution of IGC (labelled in pink colour) in
the sensitized and stressed (70% yield strength) AA2024 specimen after in situ open circuit
exposure in 0.6 M NaCl as a function of time, i.e., (a) 5.6 h, (b) 16.9 h, (c) 27.6 h, and (d) 35.9
h of exposure. Stress is applied in transverse direction, T.
Side View Top View
TL
S
300 µm
(d)
35.9 h
(c)
27.6 h
(b)
16.9 h
(a)
5.6 h
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
366
Figure 7.12 Top and side views of the three dimensional reconstructed images (about 1000
tomographic slices in each case) showing the evolution of IGC (labelled in pink colour) in
the sensitized and stressed (90% yield strength) AA2024 specimen after in situ open circuit
exposure in 0.6 M NaCl as a function of time, i.e., (a) 10.9 h, (b) 22.4 h, (c) 33.2 h, and (d)
41.2 h of exposure. Stress is applied in transverse direction, T.
300 µm
(a)
10.9 h
Side View Top View
(b)
22.4 h
(c)
33.2 h
(d)
41.2 h
T L
S
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
367
Figure 7.10 shows that in an unstressed AA2024 specimen, localized corrosion
initiates at few particular sites on the sample surface while most of the surface remains
unattacked. Figure 7.11 shows the evolution of localized corrosion in the dog-bone
AA2024 specimen stressed to 70% Y.S. Intergranular corrosion attacks initiates at the
surface and propagates in the ‘L’ direction. However, IGC growth in L direction ceases or
is limited after a certain time, but corrosion initiates at other places of the sample surface.
AA2024 specimen with 90% YS (Figure 7.12) show much more corrosion attack
compared to the unstressed or 70% stressed samples. Significant amounts of metal loss
due to the grain dissolution and/or grain fall out can be seen even after 11 h of exposure.
In most of the cases growth of the localized corrosion sites into the bulk of the specimen
seems to be followed by the general dissolution of the grain interiors and/or grain fallout,
resulting in considerable metal loss.
The effect of applied stress on the evolution of localized corrosion can easily be
seen from these figures (Figure 7.10, Figure 7.11, and Figure 7.12). Comparisons of these
three figures clearly establish that intergranular corrosion is much more severe under the
application of stress. The sample stressed to 90% YS has shown a considerable amount of
intergranular corrosion. It can be seen from the tomographs that the attack follows the
rolling direction of the alloy. The rate of growth in through thickness or short transverse
(S) direction is much slower than in the longitudinal (L) or long transverse directions (T)
directions.
The ability of X-ray synchrotron tomography technique to visualize the localized
corrosion within the bulk of opaque specimen can clearly be established from Figure
7.13, Figure 7.14, and Figure 7.15. These 3D reconstructions of the specimens are
produced by combining the information contained in a full tomographic scan data
volume, consisting of 1024 slices.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
368
Figure 7.13 Three dimensional reconstructed image (about 1000 tomographic slices in each
case) showing the morphology of IGC attacks in the sensitized AA2024 specimen (i.e., heat
treated at 250˚C for 2h) in unstressed condition after in situ open circuit exposure in 0.6 M
NaCl for 33.9 h. Intergranular corrosion has been labelled in pink colour.
Unstressed, 33.9 hT
L
S
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
369
Figure 7.14 Three dimensional reconstructed image (about 1000 tomographic slices in each
case) showing the morphology of IGC attacks in the sensitized (i.e., heat treated at 250˚C for
2h) and stressed (70% Y.S.) AA2024 specimen after in situ open circuit exposure in 0.6 M
NaCl for 35.9 h. Intergranular corrosion has been labelled in pink colour.
70% Y.S., 35.9 h T
L
S
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
370
Figure 7.15 Three dimensional reconstructed image (about 1000 tomographic slices in each
case) showing the morphology of IGC attacks in the sensitized (i.e., heat treated at 250˚C for
2h) and stressed (90% Y.S.) AA2024 specimen after in situ open circuit exposure in 0.6 M
NaCl for 22.4 h. Intergranular corrosion has been labelled in pink colour.
90% Y.S., 22.4 h
T
L
S
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
371
Figure 7.13, Figure 7.14, and Figure 7.15 are basically similar to the 3D
reconstructions already shown in Figure 7.10, Figure 7.11, and Figure 7.12. However, in
these cases the 3D views of the specimens are enlarged to establish the usefulness of X-
ray synchrotron tomographs to pick up very small details of the features present within an
object. The number, size, and morphology of all the localized corrosion sites within the
specimen can clearly be quantified from these three dimensional figures. It has to be
noted that though these pictures present the specimen view from a particular orientation,
the 3D specimen object can be rotated freely using the software to a particular orientation
of interest. In all cases the specimen interior has been made translucent to have a clear
view of the localized corrosion sites which are labelled in pink colour.
Figure 7.13 shows the 3D reconstruction of the unstressed specimen after 34 h
exposure in 0.6 M NaCl. Three predominant IGC corrosion sites can be seen in the
matchstick cylindrical specimen. Spread of these IGC sites are more in the L and T
direction than in the S direction. The specimen stressed to 70% YS (Figure 7.14) shows
similar corrosion features to that observed on the unstressed sample except that the
stressed specimen is more severely attacked. Several intergranular corrosion sites can be
seen after 36 h of exposure. Figure 7.15 shows extensive localized corrosion in the 90%
YS specimen interior after 22 h of exposure. Rapid IGC growth seems to be followed by
grain dissolution resulting in the significant material loss.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
372
7.3 Quantitative IGC Growth Rates in Unstressed and Stressed Specimens as a Function of Immersion Time
Instead of using a linear measurement of IGC growth in a single direction, this
current study quantifies the IGC damage by measuring the volume of material loss due to
the overall IGC event which includes the widening aspect of the localized corrosion.36
The amount of IGC attack (represented as volume of metal loss as a percentage of
specimen volume)37 on the unstressed and stressed specimens as a function of exposure
time in naturally aerated 0.6 M NaCl solution is shown in Figure 7.16. In the same figure,
IGC growth in sample without stress and sample with 70% YS has been emphasized in a
separate graph. It is quite obvious that stress is contributing significantly in the increase
of intergranular corrosion growth kinetics as expressed by the volume loss with
increasing exposure time. However, a huge difference in the IGC volume loss has been
noticed between the samples stressed to 90% of its yield strength and the sample stressed
to 70% of its yield strength. Intergranular corrosion of aluminium alloys are associated
with an induction period prior to the initiation [70, 95]. So, it could be possible that the
induction period is followed by initiation of IGC and rapid propagation. Due to the
limited number of exposure of each specimen under the synchrotron (4 times during the
span of almost two days), it was difficult to get more information about the initiation
stage.
36 More details about the quantitative IGC growth rate measurement in a single direction as well as IGC
damage represented by the volume of material loss can be found elsewhere [235]. 37 Depending on the geometry of the specimen (‘matchstick’ vs. ‘dog-bone’ types), exposed area under the
synchrotron may vary. So, total volume losses due to IGC from a specimen as well as the total volume of
that particular exposed specimen were calculated using the software ‘Amira’. Ratio of these two (i.e. total
volume of IGC loss : volume of the exposed specimen) are represented in the graph.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
373
0
5
10
15
20
25
30
35
0 10 20 30 40 50
No Stress70% YS90% YS
IGC
(vol
%)
Exposure Time (h)
0
0.5
1
1.5
2
2.5
0 5 10 15 20 25 30 35 40
No Stress
70% YS
IGC
(vol
%)
Exposure Time (h)
Figure 7.16 IGC in the form of volume of metal loss as a percentage of sample volume in the
sensitized (i.e., heat treated at 250˚C for 2h) AA2024 specimen in unstressed and stressed
condition after in situ open circuit exposure in 0.6 M NaCl as a function of time. It can be
seen that volume loss is significantly high in the sample stressed to 90% of yield strength.
Samples stressed to 70% YS also show higher volume loss than the unstressed sample.
Comparison of unstressed and 70% YS sample is shown separately in the inset graph.
However, IGC growth at the later stage of exposure of for all samples is presented
in Figure 7.17. It can be seen clearly that in the later stage of IGC growth both unstressed
and 70% YS samples followed similar linear growth rate relationships, whereas 90% YS
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
374
0
5
10
15
20
25
30
35
15 20 25 30 35 40 45
No Stress70% YS90% YS
y = -0.23393 + 0.023685x R= 0.99946
y = 0.031185 + 0.060152x R= 0.94957
y = -10.187 + 0.9973x R= 1
IGC
(vol
%)
Exposure Time (h)
sample is markedly different with much higher slope in the linear growth rate. After 40
hours of exposure, the 90% YS sample lost almost 30% of its volume due to intergranular
attack.
Figure 7.17 IGC in the form of volume of metal loss as a percentage of sample volume at the
later stage of exposure of the sensitized (i.e., heat treated at 250˚C for 2h) AA2024 specimen
in unstressed and stressed condition during in situ open circuit exposure in 0.6 M NaCl. It
can be seen from the best fit of the curves that IGC growth follows a linear relationship with
time. The slope for the 90% sample is much higher than the unstressed and 70% YS sample
indicating higher amount of material loss due to IGC with higher stress.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
375
7.4 Discussion
The advantages of the X-ray synchrotron technique over the conventional 2D
techniques have been discussed in Section 2.4 of Chapter 2 and in the beginning of this
chapter. The results from this current study not only demonstrate different
features/information that can be obtained from the bulk material, but it also shows the
enormous potential of the X-ray synchrotron tomography as a very useful non-destructive
technique. This study also establishes the effect of applied stress on localized corrosion
(predominantly intergranular corrosion) propagation in aluminium alloys. Standard
electrochemical tests and subsequent microstructural analysis of the exposed samples can
provide information about the susceptibility of a material towards localized corrosion.
However, as most of these conventional techniques are based on observations in two
dimensions, they can not provide information about the true shape, connectivity,
morphology and sizes of the localized corrosion sites. Information about the development
of such localized corrosion sites as a function of exposure time or its interaction with
microstructural features can not be obtained either. Hence, difficulties remain with the
conventional 2D techniques to accurately measure localized corrosion growth rates.
This current study demonstrates that X-ray synchrotron tomography can be used
to visualize and characterize the development and quantify the growth rates of localized
corrosion sites. High voxel38 resolution (~ 0.7 µm) in the current study allows the
possibility of very detailed and accurate study of the interior of the specimen in both two
and three dimensions. Analysis software also allows the flexibility to view a particular
specimen in three dimensions (or a single slice of a specimen at a particular plane in two
38 A voxel is a three dimensional pixel.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
376
dimensions) at any position and any orientation. These flexibilities associated with this
technique help in monitoring and quantifying the localized corrosion growth in S, L or T
directions. However, it should be noted that human error during the use of the software
(Amira) can result into the inaccurate quantitative analysis. During 3D data analysis using
the software, voxels of different features are labelled from their respective contrast
values. As the threshold values used to differentiate these regions are set by the person
performing the analysis, any human misjudgement could lead to the inaccuracy of the
analysis [235].
Apart from these, the limitations of the specimen size and spatial resolution during
the synchrotron experiments could potentially result into the underestimation of growth
rates. As the X-ray tomographic technique depends on the transmission of the beam
through the specimen, energy used during the experiment often limits the size of the
specimen to be used. With beam energy of 17.5 keV, 500 µm diameter specimens are
used in this current study to ensure sufficient transmission of the incident X-ray beam.
Small sample size may affect the penetration kinetics in a particular direction. For
examples, in many cases during the current study IGC is seen to be rapidly penetrated the
thickness of the specimen in the L direction. So, it is quite possible that with larger
diameter of the specimen further growth would have been observed. It is also speculated
that small sample size can limit the amount of anodic dissolution by providing smaller
surface area for supporting cathodic reaction [235].
As described earlier (Chapter 2), aluminium alloy 2024 is characterized by the
presence of intermetallic particles which have high atomic numbers compared to the
matrix and as a result they appear white in the tomographs with significant contrasts. It is
believed that these intermetallic particles are associated with the initiation sites of
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
377
localized corrosion in the alloy [8].39 So, detailed investigations have been performed
earlier at the University of Birmingham to obtain a better understanding on how the
presence of intermetallic particles affects the propagation of intergranular attack [235,
288]. Connolly et al. [288] described the appearance of different features in a
tomographic slice of sensitized AA2024 in great detail. The main features of a particular
tomographic slice (Figure 7.1) of this current study is consistent to that reported by
Connolly et al. [288]. It is quite clear from these tomographic images that X-ray
tomography can be used accurately to image and measure different features like
intermetallic particles which are aligned in the rolling direction, pre-existing cavities in
the matrix and adjacent to the particles, localized attack and their interactions with the
microstructural features etc.
3D reconstructions of some corroded areas from both unstressed and stressed
specimens (Figure 7.8) along with the after-exposure SEM analysis of the unstressed
specimen (Figure 7.9) suggests the intergranular nature of the localized attack during this
current study. All attacks in this current study (in both stressed and unstressed sensitized
temper specimens) are found to be intergranular in nature with absence of any cracking
even after 36h of exposure in this particular environment (0.6 M NaCl). Intergranular
corrosion is a type of localized attack at and adjacent to the grain boundaries as a result of
an enhanced thermodynamic and/or kinetic tendency for corrosion relative to the grain
matrix [111, 217]. IGC is operated by a pre-existing active dissolution path based on
compositional/structural heterogeneity. In AA2024 intergranular corrosion is believed to
start with the dissolution of the S phase particles (Al2CuMg) on the grain boundaries
39 Detailed discussion about the role of intermetallic particles in the initiation of localized corrosion can be
found in the literature review (Chapter 2) and in Chapter 5 (Microelectrochemical studies on single
intermetallic particles).
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
378
caused by the galvanic coupling between the solute-depleted zone/surrounding matrix and
solute-depleted zone/S phase particles [7]. Subsequently, IGC propagates by the
dissolution of the copper depleted zone along the grain boundaries [95].40
Anisotropic growth of localized corrosion in AA2024-T3 has been confirmed by
several researchers and it was also found that the 2D penetration rate of localized attack
depends on the direction of the applied stress [29, 52, 233, 319]. So, it is important to
know the orientation of the sample under examination to have a proper idea about the
localized corrosion growth in three dimensions. Figure 7.2 shows the schematic of a
typical specimen with these three different orientations. Stressed samples were subjected
to load in the transverse (T) direction. Figure 7.2 also schematically explains how a single
slice from a single plane is represented within the bulk specimen.
In both unstressed and stressed specimens, IGC attacks are found to be propagated
in the L direction at a faster rate (Figure 7.4- Figure 7.6). As the L direction in the
specimen is also the rolling direction of the source plate, it is possible that IGC attacks are
following the easier paths of longer grains. Spread of the IGC attacks in the S direction is
smaller than in the L and T direction. These observations are consistent with the other
researchers [233] who reported slow growth rate of IGC in S direction as a result of
longer path resulting from ‘squeezed’ grains in the short transverse direction. It is also
evident from the 2D tomographs that localized corrosion does not necessarily initiate
from the intermetallic particles every time and is therefore consistent with the findings of
others [235]. However, the amount of localized corrosion attack in the presence of stress
is much higher than in the unstressed condition. In some cases, severe undercutting of the
stressed specimen surfaces have been seen (Figure 7.6b and c). This may be an indication 40 Mechanisms of intergranular corrosion in aluminium alloys and its relation with different heat treatment
could be found in Section 2.2.2 of Chapter 2.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
379
that the machining process for manufacturing the specimens can actually affect the
corrosion susceptibility and the rate of local attacks on the surface [288].
As seen in unstressed samples, IGC in stressed samples also stops growing (or the
growth rate significantly decreases) in the L direction and broadening of IGC occurs
through grain interior dissolution and/or grain fall out. Volume analysis in three
dimensions (3D) [Figure 7.10, Figure 7.11, and Figure 7.12] showed an increase in
volume loss as a function of time which proves that, though IGC stop growing in L
direction, it continues to grow in other directions. It is very possible that for both stressed
and unstressed samples many localized corrosion sites might initiate as pitting attacks
which later develop as intergranular corrosion.
As discussed earlier, the localized corrosion attacks in the current study seems to
follow the elongated grains in the rolling direction (see Figure 7.8). Prolonged exposure
in an aggressive solution will lead to the coalescence of these attacks and forming of a
plate like IGC network in the transverse direction as can be seen in the side view from
Figure 7.8 (more elaborative view of the IGC network within a specimen interior can be
seen in Figure 7.13, Figure 7.14, and Figure 7.15). The nature and morphology of
intergranular attacks in this current study with and without application of stress are very
similar to that reported by Fox [235] in unstressed sample under similar experimental
conditions. These similarities emphasize the fact that stress is actually increasing the
intergranular corrosion kinetics rather than promoting cracking as such.
The influence of mechanical/applied stress on the kinetics of intergranular
corrosion is complex and has been debated. Stress corrosion cracking is driven by the
synergistic actions of mechanical and electrochemical factors. So, it is thought that IGC
and IGSCC are closely linked phenomena, and it has been viewed by some researchers
that IGSCC is nothing but stress assisted IGC [29]. However reported experimental
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
380
results are contradictory and can be divided into main groups like: (a) stress induced
intergranular corrosion, and (b) stress assisted intergranular corrosion [218]. When in a
mechanically stressed specimen, intergranular attacks occur under less aggressive
electrochemical conditions compared to the unstressed condition, it is known as stressed
induced intergranular corrosion. By comparison, if under identical electrochemical
conditions, the growth rate of intergranular attacks is higher in mechanically stressed
specimens, it is known as stress assisted intergranular corrosion. According to this
convention, as both unstressed and stressed specimens in the present study are in identical
electrochemical condition and application of stress seems to increase the kinetics of
intergranular corrosion, the observed attacks in the stressed sample in this current study
can be classified as stress assisted intergranular corrosion.
As the accurate IGC rates are difficult to quantify using conventional two
dimensional measuring methods, the X-Ray tomography is probably one of the most
valuable tools to quantify IGC growth rate in 3D. Earlier studies by the foil penetration
technique [29, 233, 320] showed that growth rate of localized attack was anisotropic
because the attack was intergranular corrosion (IGC) and the grains were elongated in the
rolling direction. In the foil penetration technique, the growth kinetics of localized
corrosion is measured by the time required to penetrate a thin foil [29, 49-52]. Liu et al.
[29, 319] studied the effect of tensile stress on the intergranular corrosion of AA2024-T3
using a modified foil penetration technique. This group also used X-ray radiography41 to
image in situ initiation and growth of multiple IGC sites in AA2024 in unstressed
41 X-ray radiography captures the 2D radiographic projection of the samples whereas in X-ray tomography
a series of recorded 2D radiographic projections (taken while the sample is being rotated from 0˚ to 180˚)
are used to reconstruct a 3D map of the X-ray attenuation coefficient of the material using appropriate
algorithm.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
381
condition as well as under the application of normal tensile stress [53-55]. It was found
that the penetration rate of corrosion attack depends on the applied potential and stress
state with samples showing higher penetration rate when subjected to an applied load.
However, the main disadvantage with the foil penetration technique is that it
determines the growth kinetics of the fastest growing localized corrosion sites. In this
present study it has been observed several times that IGC growth might cease or be
limited in a particular direction (e.g., in many cases IGC attacks stops growing in the L
direction), but continues in other directions. These observations emphasize the limitations
of foil penetration technique. Increase in the IGC rates in the presence of a tensile stress
at the surface of the material was also reported by Rota et al. [218]. Zhao et al. [55]
studied the intergranular corrosion of aluminium alloys using X-Ray radiograph and
found that growth kinetics determined from radiographs of the samples were slower than
growth kinetics determined by foil penetration technique. In a separate study Zhao and
Frankel [35] investigated the effect of prior deformation on localized corrosion of
aluminium alloys. AA2024-T3 showed increased corrosion rate after pre-stressing, but
little change in breakdown potentials. It is speculated that the increase in the current
density with increase in pre-strain is due to the deformation induced defects.
It was also found that the nominal growth rate in the short transverse (S) direction
is slower because of the longer path length around the grains in that direction.42 It is
confirmed from this study that corrosion propagates at a much faster rate in the L and T
direction in both stressed and unstressed specimen. The distribution of IGC in the T
direction (which is also the stressing direction) can be seen from the three dimensional
42 Refer back to Figure 2.23 (Chapter 2) for grain structure in different directions of AA2024.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
382
rendering of the samples, Figure 7.10 - Figure 7.12. IGC attacks seem to be initiated from
the edge of the surface and propagate in the L direction until a certain depth.
Rapid growth of IGC to a certain depth and its subsequent cease or limit in growth
rate could possibly be explained by the self-limiting effect of localized corrosion as
discussed by Lifka [221]. IGC initiate and grow rapidly to a certain depth at which it
becomes mass transport limited. Growth of the IGC attacks then decrease significantly
since it becomes more difficult for the oxygen and the corroding species to transport
down to the active tip of the IGC site through the narrow and tortuous corrosion path.
Severe branching of IGC networks make it extremely difficult for oxygen to diffuse down
to the active tip to provide necessary cathodic support for continuing anodic dissolution.
Application of stress is thought to play an important role at this stage. Stress would tend
to rupture any film that might form at the IGC tip and also effectively open up the IGC
path, thereby increasing mass transport and decrease the ohmic potential drop down the
narrow IGC path [29, 218]. When penetration of IGC in a certain direction ceases, it
begins to spread laterally through the material in other directions [221].
This theory has been experimentally supported by the three dimensional
visualization of IGC by Fox [235], who found that IGC sites rapidly penetrate a
susceptible alloy to the limiting depth at which point the electrochemical driving force for
corrosion would start favouring the grain interiors rather than the grain boundaries. At
this point widening of IGC will occur as found in this current study and also observed by
other researchers [52, 321]. Augustin et al. [321] studied the kinetics of intergranular
corrosion in AA2024-T351 by immersing the samples in 1 and 3 M NaCl followed by
optical and SEM observations on the sectioned samples. Augustin found that the
intergranular corrosion corresponds not only to an increase in the depth of the corrosion
attacks but also an increase of the width of the attacks. During their study, depth of
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
383
penetration of IGC attacks in AA2024-T351 remained the same between the exposure
time of 72 and 168 hours; however, the IGC attacks of the 168 h immersed samples were
much wider compared to the 72 h exposed samples. The dissolution of the interior grains
could also result due to the possible change in chemistry inside a dissolving boundary.
Kotsikos et al. [236] suggested that aggressive ion accumulation inside an active
tip of an IGC can occur within a few hours. This increase in aggressiveness could play an
important role in spreading the dissolution from the boundaries to the grain interiors.
Though growth ceases or is limited in the L direction, the localized corrosion sites start
widening up in the S direction as well as continuing their growth in the T direction.
Several active sites in the T direction from the outer surface of the exposed sample would
then coalesce, thereby contributing to the volume loss of the material. Propagation of IGC
as a function of time in the S direction is slow and that can be observed from the 3D and
2D figures. Grains fall out from the matrix as the IGC attack becomes significant and
surrounds a grain in three dimensions. Wedging stress generated by the corrosion product
inside the IGC attack may also play an important role in assisting the IGC process [29,
53]. Local wedging stress from the corrosion product could be significant enough to
prevent repassivation in the localized corrosion environment.
Detailed quantification of the growth rate of IGC in aluminium alloys has been
limited in the open literature. In most of the studies growth rate has been measured as 2D
penetration in different orientations.43 IGC growth is normally found to follow a tn
penetration relationship which is represented as d = Ktn, where d is the depth of
penetration, K is a constant, t is time and n is a fraction between 0 and 1 [29, 55, 319,
320]. However, it should be noted that in most of these cases, the test specimens were
43 See Section 2.2.2.1 (Chapter 2) for more details about Growth Kinetics of IGC in Aluminium Alloys.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
384
polarized above the pitting potential of the alloy. Hence, it is possible that under those
experimental conditions grain boundary dissolution is accelerated and therefore the IGC
rates measured are not really representative of what an alloy might encounter in real
service life. It is interesting to note that, when in this current study the intergranular
corrosion growth rate is measured in three dimensions and represented as volume loss as
a function of exposure time, the growth rate seems to follow a linear relationship (see
Figure 7.17).
Rota and Boehni [218] speculated that the effect stress on the growth kinetics of
localized corrosion is strongly dependent on the number of parallel growing
attacks/cracks in the exposed area. Intergranular corrosion growth of AA2024 under
applied stress was divided in three subsequent stages: (i) activation, (ii) transition, and
(iii) stable, macroscopic grain dissolution stage. Effect of stress was found to be only in
the first stage where a small number of parallel cracks/localized attacks were active. In
the second stage those localized attacks starts forming network and branching, thereby
making the effect of stress less predominant. In the third stage, grain dissolution starts and
growth kinetics become independent of applied stress. Rota tried to explain the effect of
stress based on stress induced widening of the attacks which not only reduces the integral
ohmic resistance of the system but also improves the mass transport rate. With the
generation of large number of parallel cracks/localized attacks and network formation,
chances of stress induced widening of IGC attacks diminished. This phenomena result
into the disappearance of stress effects in the presence of large number of parallel IGC
attacks. However it must be noted that Rota’s model of stress induced IGC growth
kinetics is based on the results obtained from a modified foil penetration test set up which
measure the fastest growing localized attack in two dimensions.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
385
Observations of the localized corrosion growth kinetics in two dimensions in this
current study are quite similar to that reported by Rota and Boehni [218] in the stressed
and unstressed conditions. Initial rapid growths of the intergranular attacks in the L
direction have been found for both stressed and unstressed sample (Figure 7.4 - Figure 7.6
and Figure 7.10 - Figure 7.12) before reaching self-limiting growth. However, in all the
cases grain dissolution continues as a function of immersion time as seen from the three
dimensional volume analysis (Figure 7.16). The effect of stress in the initiation stage44 of
IGC can be confirmed by comparing unstressed and 70% YS samples. With the
application of stress, IGC volume loss is rapid during the initial immersion. Due to the
limited number of in situ observations of each specimen under the synchrotron (4 scans
on individual specimens during the span of almost two days), it was difficult to get more
information about the initiation stage. During the later stage of immersion (15h – 36 h),
growth kinetics as expressed by the volume loss is similar for both the unstressed and
70% stressed samples (Figure 7.17).
It should also be noted that though the total current remains constant as a function
of time during the third stage of IGC growth reported by Rota and Boehni [218], the
amount of charge passed which is a direct measure of total volume loss still increases
with increasing exposure time. So, in this present study it is not surprising to see that the
volume loss due to IGC increases during the later stage of exposure. A significant
difference in the volume loss between the 70% YS sample and 90% YS sample has been
found in this study. After 35 hours of exposure, IGC volume loss in the 90% YS sample
is almost one order of magnitude higher than that of the unstressed or 70% YS sample.
44 Effect of applied stress on the initiation of localized corrosion is discussed in Chapter 6. It was found that
applied stress equivalent of 70% YS or more had an adverse effect on the corrosion properties of AA2024-
T351.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
386
The exact reason for this significant increase in the volume loss for a sample stressed to
90% of its yield strength is not properly understood as yet. But it is possible that once the
microstructure is weakened by the IGC growth, the grains starts to slide apart more
rapidly in a 90% YS sample (as it is very closed to the plastic domain of the material)
than in a unstressed or in a 70% YS sample. These kinds of sliding will open up the
grains along their edges and once these grain edges coalesce, IGC propagates through the
microstructure. It is also possible that the specimen stressed to 90% YS starts off with
many more initiation sites compared to the unstressed or 70% YS specimens. Higher
number of such initiation sites would contribute to the higher volume of material loss in
the later stage.
As it has already been discussed in Chapter 6,45 during potentiostatic polarization
of AA2024-T351 in aerated 0.01 M NaCl in the anodic region (between OCP and the
breakdown potential), the difference in the charge passed (which is directly proportional
to the metal dissolution) between an unstressed sample and a sample stressed to 45% of
its yield strength is negligible (see Figure 6.34 in Chapter 6). When the sample is stressed
to 70% of its yield strength, slight increase in the passed charge occurs (see Figure 6.37 in
Chapter 6) and when the applied stress is 90% of the yield strength, charge passed for
stressed sample is always much higher in comparison to the unstressed sample (see
Figure 6.40 in Chapter 6). This possibly emphasizes the fact that, microstructural changes
in a sample could happen close to its yield strength, and this might have an adverse effect
on corrosion susceptibility.
Application of stress is found to decrease the initiation time for localized
corrosion in stainless steel without increasing the number of nucleation sites significantly 45 See Chapter 6 to find out more about the effect of elastic and plastic stress on the localized corrosion of
aluminium alloys.
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
387
[322]. Application of stress is believed to generate surface defects such as slip lines, slip
bands, microcracks, decohesion between phases [29, 35, 36, 323, 324]. All these surface
defects may act as potential sites for corrosion initiation. However, individual
contribution from each of these defects or possible synergistic effects of more than one
factor still remains a subject of further investigations.
7.5 Conclusions
1. X-ray synchrotron tomography is a powerful non-destructive technique which can
be used to visualize, monitor, and quantify real time corrosion development in
three dimensions within the bulk of a material in micron and submicron scale.
This technique has been shown to be an effective tool for better understanding of
the electrochemical and physical mechanisms responsible for the development as
well as propagation of localized corrosion sites (i.e., intergranular corrosion in this
case) in aluminium alloys.
2. Quantitative growth rates of intergranular corrosion along with their
morphological evolution in unstressed and stressed condition provide valuable
input that can have significant contribution in the life prediction modelling tool of
the aircraft components.
3. This current study documents the effect of stress on the growth rate of
intergranular corrosion for the first time in three dimensions. It has been shown
that 2D technique may not always quantify the growth rate accurately as localized
corrosion propagation can stop in a particular direction but still continuing growth
Chapter 7. X-ray Synchrotron Tomographic Study of Localized Corrosion Propagation in Aluminium Alloys
388
in other directions. Under these circumstances, 3D analysis has been seen to
provide a better description of the corrosion process.
4. Application of stress of 70% YS or above has a significant effect on the localized
corrosion propagation of the sample. Increase in the remotely applied stress
increased the amount of attack.
5. All attacks were intergranular corrosion in nature, no cracking was observed even
after long exposure in this particular environment. 3D reconstruction provides the
visual evidence as to how IGC propagates in three dimensions.
6. Growth of the IGC attacks in the L direction stop after certain time in most of the
samples.
7. A linear relationship between the volume loss due to IGC and exposure time has
been observed in both stressed and unstressed sample. Sample stressed to 90% of
its yield strength showed rapid increase in the volume loss due to IGC with
increasing immersion time and resulting volume loss was almost 30% of the total
sample volume.
Chapter 8. Overall Summary and Conclusions
389
8 OVERALL SUMMARY AND CONCLUSIONS
In this project, efforts have been made to gain a better understanding of the role
of intermetallic particles and applied stress in initiating and propagating localized
corrosion (mainly pitting and intergranular corrosion) in aluminium alloy 2024 (Al-Cu-
Mg). The initiation stage of the localized corrosion was investigated using macro and
micro-scale electrochemical techniques whereas propagation of localized corrosion was
studied with X-ray synchrotron tomographic techniques.
8.1 Effect of Stress on Localized Corrosion Initiation
In earlier studies by other researchers (in unstressed conditions), initiation of
localized corrosion in aluminium alloys has been associated with the presence of
intermetallic particles on the alloy surface [8-11, 14, 15, 37, 38, 40, 42, 47, 48, 82, 123,
171, 173, 175, 181-197]. However, the role of those intermetallic particles in initiating
localized corrosion has not been well studied under stressing conditions. It has also not
been well documented whether morphological changes in the intermetallic particles play
any key role in controlling the corrosion properties of the alloy under stress.
The role of applied stress in the initiation of localized corrosion was investigated
with a special emphasis on the behaviour of intermetallic particles under stress.
Application of plastic stress on AA2024-T351 was found to decrease the corrosion and
breakdown potentials of the alloy. It has been seen in this study that application of stress
creates delamination at the intermetallic particle/matrix interface. Presence of any
Chapter 8. Overall Summary and Conclusions
390
micro/nano crevices (which could be produced by particle/matrix delamination during
application of stress) at the particle edge are thought to decrease the breakdown potential
of the alloy by increasing the electrochemical reactivity at those tight creviced areas.
It should be noted that most of the particles do not show any delamination even at
a high plastic load (equivalent to 140% of the yield strength of AA2024-T351). However,
due to the presence of a large number of intermetallic particles in AA2024-T351, the
likelihood of exposing such a flaw is high even with a very small exposure area. Since the
localized corrosion properties of an alloy could be determined even by a single weak
point, it is highly probably that the exposed area of even one mm diameter would contain
such delaminated areas. In a few cases during the current study, the application of plastic
stress does not change the breakdown potential of the AA2024-T351 specimens. These
observations are probably indicative of the fact that the corrosion performance of the
alloy might be controlled by the probability of the existence of delamination at the
particle/matrix interface in the exposed areas. Electrochemical experiments on the surface
treated AA2024-T351 samples (i.e., without ‘S’ phase particles) do not show total
elimination of the stress effect (as the breakdown potential of the alloy is still lowered by
the application of plastic stress). These observations emphasize the fact that the presence
of intermetallic particles other than ‘S’ phase particles could also play a role in
determining the corrosion behaviour of the alloy under stressing condition.
Further electrochemical experiments with an intermetallic particle free Al-
0.099Cu binary alloy also points out at the possible crucial role of the presence of
intermetallic particle under stressing conditions. The binary alloy (Al-0.099Cu) without
any intermetallic particles does not show any drop in breakdown potential after
application of plastic stress though it does show drop in corrosion potential. The
plastically stressed intermetallic particle free Al-0.099Cu binary alloy shows the presence
Chapter 8. Overall Summary and Conclusions
391
of slip lines/bands and related microcracks on the alloy surface and these features are seen
to interact with the electrolyte during the corrosion experiments. However, the absence of
any change in the breakdown potential of the plastically stressed Al-0.099Cu binary alloy
during the corrosion experiment indicates that in the absence of intermetallic particles, the
slip bands or rough surface may not be the main responsible features in changing the
breakdown potential of the alloy.
The importance of the delamination at the particle/matrix interface as the possible
key controlling factor on localized corrosion initiation in AA2024-T351 has further been
emphasized by the experimental results from micro-capillary electrochemical tests. It has
been found that application of plastic stress close to the ultimate tensile strength of the
alloy does not significantly change the corrosion behaviour of the particle free matrix.
The results also indicate that in the absence of a delamination at the particle/matrix
interface, plastically stressed Fe-Mn rich particles do not exhibit any change in the
breakdown potential.
Since it seems that formation of a tight crevice at the particle/matrix interface can
potentially control the localized corrosion initiation, it is logical to assume that any
process that minimises the possibility of a tight crevice formation could actually increase
the breakdown potential. This argument could possibly be supported by the observed
higher breakdown potentials of the Fe-Mn particles with longer cathodic polarization
during the micro-capillary electrochemical experiments. Longer duration of the cathodic
polarization will generate more alkalinity which may create alkaline grooving around the
particles. These relatively wider grooves can alter the geometry of tight crevices and
thereby minimize the chance of initiating pits. Presence of such alkaline grooving around
the particles was verified through optical profilometry and AFM analysis in this current
study.
Chapter 8. Overall Summary and Conclusions
392
8.2 Effect of Stress on Localized Corrosion Propagation
As in most of the practical applications, aluminium alloys are subjected to stress
in order to achieve a specific structure; researchers try to understand the effect of stress
on the growth kinetics of localized corrosion. Inherent limitations of the two dimensional
techniques that are used to investigate the anisotropic growth rate of localized corrosion
in aluminium alloys impose difficulties in knowing the actual growth rate in a real bulk
material in three dimensions. Two dimensional analysis of the corrosion growth also do
not provide any information about the true shape, connectivity, and morphology of
corrosion sites along with their interactions with different microstructural features.
This current study emphasizes that X-ray synchrotron tomography can be used to
visualize and characterize the development and growth of localized corrosion quite
accurately in three dimensions. Hence, it is demonstrated here that X-ray synchrotron
tomography can overcome the limitations associated with the conventional two
dimensional techniques. This study also establishes the effect of applied stress on
localized corrosion (predominantly intergranular corrosion) propagation in aluminium
alloys. It has been found that application of stress of 70% Y.S. or above has a significant
effect on the localized corrosion propagation within the sample. An increase in the
applied stress increases the amount of attack. All attack sites are found to be intergranular
in nature and no cracking is observed even after long exposure in the environment of
naturally aerated 0.6M NaCl at room temperature.
3D reconstruction and rendering of X-ray attenuation data visually represents the
propagation of intergranular corrosion in three dimensions. The propagation rate of the
intergranular attack is found to be faster in the rolling (L) and in transverse (stressing)
Chapter 8. Overall Summary and Conclusions
393
direction compared with the short transverse direction. In the short transverse (S)
direction, the growth rate was slower probably because of the longer path around the
grains in that direction. It is interesting to note that using two dimensional techniques,
several researchers found the intergranular corrosion growth to follow a tn penetration
relationship which is represented as d = Ktn, where d is the depth of penetration, K is a
constant, t is time and n is a fraction between 0 and 1 [29, 55, 319, 320]. However, in this
current study, representing the growth rate of intergranular corrosion as the volume of
material loss as a function of exposure time, a linear relationship has been observed in
both stressed and unstressed samples. A sample stressed to 90% of its yield strength
shows a rapid increase in the volume loss due to intergranular corrosion with increasing
immersion time and results in almost 30% loss of the total sample volume. Thus, this
current study shows the effect of stress on the growth kinetics of localized corrosion in
three dimensions for the first time in aluminium alloys.
Chapter 9. Future Work
394
9 FUTURE WORK
Modelling of corrosion of aluminium alloys has become a subject of great interest
to the aircraft manufacturers and maintainers during the last decade or so. Fundamental
understanding of the corrosion initiation process in aluminium alloys will be helpful to
develop better mitigation strategies and inspection schedules. Moreover, techniques
which provide better insights and subsequent model inputs into the growth kinetics of
localized corrosion will enhance the success of these predictive models with higher
accuracy.
Different theories have been proposed to explain the effect of stress in changing
the corrosion and breakdown potentials in aluminium alloys, but none are complete on
their own. Moreover, very limited research has been carried out on the intermetallic
particle behaviour under applied stress and its subsequent influence on corrosion
properties. In this work, experimental evidence has been found which indicates that
delamination at the particle/matrix interface can provide corrosion initiation sites by
creating nano/microcrevices. However, considering the fact that delamination does not
occur in all the particles and AA2024 has about 300,000 particles in every inch square,
statistics seems to play a significant role in characterizing the localized corrosion
initiation. Hence, some statistical studies should be performed to calculate the percentage
of particle delamination and the probability of delamination for a given surface area and
then correlate these calculations with observed corrosion behaviour.
Micro-capillary electrochemical cell testing in conjunction with the stressing stage
provides a good opportunity to extend the statistical based study on individual
intermetallic particles of AA2024-T351. Further micro-capillary electrochemical
Chapter 9. Future Work
395
experiments can be performed on additional (i.e., on the order of 100 to 1000 for
statistical significance) individual particles under stressing condition to get quantitative
evidence for the delamination-induced localized corrosion initiation theory. These
experiments can also be used to understand the role of elastic and plastic stress in
controlling the corrosion properties in aluminium alloys. It is also possible to expand
these test matrixes to other alloy systems used in structural components for better
understanding of the local electrochemical behaviour of second phase particles, inclusions
etc. in both stressed and unstressed condition.
The latest generation of X-ray synchrotron microtomography beam lines have
been shown to be very powerful tools to the materials and corrosion scientists. 3D
information characterising localized corrosion has been obtained from the bulk of the
material using this non destructive technique. Growth kinetics has been measured as a
function of material loss in both unstressed and stressed samples. However, additional
experiments should be performed to provide additional data for a better and accurate
understanding of the growth rates. It has to be mentioned that, due to the lack of beam
line time, experiments could only be performed on single specimens in the unstressed
condition and in specimen stressed to 70% and 90% Y.S. Though this current study can
be taken as a proof of concept demonstration showing the possibility of getting new
information using X-ray synchrotron tomography, the results obtained from this study
should be verified with repeat experiments to ensure consistency. The growth law
obtained utilising the total volume percent of corroded material as the metric for growth
rate should also be verified with repeat experiments in both stressed and unstressed
samples. Moreover, due to the capabilities of the current synchrotron facilities, gaps exist
in the recorded evolution of the localized corrosion events as attack occurs between two
consecutive tomographic scans. Hence, more experiments with shorter intervals between
Chapter 9. Future Work
396
two consecutive scans should be performed utilising the newest synchrotron technology
and thereby enhancing the chance of detecting the corrosion initiation and documenting
the propagation phenomena more accurately.
As the scientists on the X-ray synchrotron beam-line improve the resolution of the
tomographic scans, the ability to perform in situ localized corrosion tests with higher
resolution (i.e., in sub micron level) on the aluminium alloys in both stressed and
unstressed conditions would be recommended. These experiments would help to get
statistical data on the particle delamination. Micro-capillary electrochemical tests could
also be performed in conjunction with stressing stage under the synchrotron beam. Faster
scan rate and high resolution of X-ray synchrotron technique can make it possible to
directly correlate the localized corrosion initiation phenomena with responsible
microstructural features.
Appendix A: Mechanisms and Models for Pit Initiation
397
APPENDIX A: MECHANISMS AND MODELS FOR PIT INITIATION
A.1 Adsorption Mechanism
Adsorption mechanism for pitting corrosion initiation is based on the reversible
competitive adsorption of aggressive ions like Cl- with oxygen for sites on the metal or
alloy surface [18, 112, 129, 130, 135, 155]. Pits develop at sites where passivating oxygen
adsorbed on the metal surface is displaced by aggressive anions like halides. The
adsorbed halides induce pitting by weakening the metal ions to metal lattice bond or by
thinning the passivating oxide film [134].
Hoar et al. [135] proposed that halide ions gets adsorbed on the oxide film of the
surface and form traditional complexes which will immediately separate from the oxide
ions in the lattice and dissolve in the solution. The rate of dissolution of the cations in the
form of a complex is much higher than the non or aquo-complexed cations present in the
film surface in the absence of halide ions. Once a cation is dissolved in the solution,
another cation comes through the film to replace the dissolved cation. However, after
reaching the film/solution interface the cation does not find any stabilizing oxide ion,
rather it finds several halide ions. Hence, once begun, this process has a strong probability
to be ‘catalytic’ in nature and thereby causing localized breakdown of the film. Apart
from halide ions, depending on the size and charge of the ions, other ions may also
penetrate the passive oxide by contaminating it and making it a better ionic conductor.
However, according to Hoar’s adsorption model, anions other than Cl-, e.g. SO42-, ClO4
2-,
NO3-, OH- should also accelerate pitting by contaminating the oxide film. However, it is
Appendix A: Mechanisms and Models for Pit Initiation
398
found that SO42-etc. is inhibitor to the pitting attack. Inhibition even found to occur at
potential where pitting would normally occurs in the absence of those ions.
So, to explain these discrepancies, Leckie and Uhlig [130] proposed an alternate
model based on the competitive adsorption of the oxygen and chloride ions. Adsorbed
oxygen rather than metal oxide is considered to make up the passive film. When O2- is
adsorbed, the metal passivates whereas adsorption of chlorides does not produce a passive
surface. Thus, above a critical potential where Cl- adsorption is favoured over O2-
adsorption, localized breakdown of the passivity occurs. When anions other than Cl- are
present, these also tend to adsorb on the passive metal/alloy surface displacing the Cl-.
Hence, in the presence of these ions, the competitive process requires a shift of potential
to still more noble values where Cl- concentration in the double layer would be adequate
to displace the oxygen from the passive film and thereby destroying the local passivity
[130].
The aspects of the adsorption model are still relevant, though it is now known that
the passive film is at least several mono layers thick rather than just an adsorbed oxygen
layer [112]. It has been found that chloride and other halides can cause thinning of the
passive film even under condition where as pit had not formed.
A.2 Penetration Mechanism
According to this mechanism, aggressive anions such as Cl- incorporate into the
passive film and migrate through it. Breakdown of passivity occurs when the aggressive
anion reaches the underlying metal [18, 112, 128, 132, 137]. Migration of anion through
the passive film would be assisted by the presence of high electric field in the film.
Appendix A: Mechanisms and Models for Pit Initiation
399
The existence of an induction time prior to the pitting process supports the idea
behind the penetration mechanism. However, it should be noted that the theoretically
calculated time for Cl- penetration through the oxide film is much longer than the
induction time measured experimentally. A critical concentration of chloride in the inner
oxide portion is thought to be associated with film breakdown and pit initiation [112,
155].
Penetration models could be divided into two broad categories: (i) according to this
type, penetration of chlorides occurs through the imperfections in the passive film, and
(ii) in the second type, penetration is believed to occur with some interaction of chloride
with the oxide lattice [18].
According to Hoar et al. [325] anions enter the oxide film under the influence of an
electrostatic field across the film/solution interface. Initiation of pit occurs when the field
reaches a critical value corresponding to the breakdown potential. Presence of
imperfections (e.g., grain boundaries) in the passive film makes the entrance of the
aggressive anions much easier. Efficiency of those anions in breakdown depends on their
respective ease of entrance through the passive film. The smaller anions like Cl- penetrate
more readily and hence more aggressive than the Br- and I-. It is also suspected that oxide
film ‘contaminated’ with aggressive anions are better ionic conductor than the original
passivating oxide [325].
McCafferty [120] supported the second type of penetration model where it is
considered that penetration of chloride ions can occur through the oxide film dissolution.
According to this model chloride ion gets adsorbed on the oxide surface, mostly on the
positively charged region of the surface. This model has been extended in such a way that
the anion penetration through the oxide film could be explained by oxide film dissolution
as well as by migration through the oxygen vacancies.
Appendix A: Mechanisms and Models for Pit Initiation
400
On the other hand, Mattin and Burstein [326, 327] proposed a model where both
chloride and oxide ions are suggested to be drawn through the passivating oxide film
under the influence of a high electric field. Oxide ions react at the metal/film interface to
form passivating metal oxide, whereas chloride form metal chloride after reaching at that
interface. Formation of solid metal chloride at that interface would expand the interface
because of the higher molar volume of the metal chlorides than the metal oxide or the
metal itself. This expansion causes mechanical rupture of the oxide film leading to a
microscopic explosion which could be the precursor of the pit nucleation [18].
A.3 Film Breakdown Mechanism
Pit initiation by film breakdown considers that the thin passive film is in a continual
state of breakdown and repair [18, 112, 137]. This model assumes the existence of
competition between film formation and metal dissolution. Pitting thought to occur at
potentials where the rate of passivity breakdown is greater than that of repassivation
[134]. According to Smialowska [18] the film breakdown mechanism (also termed as
“depassivation-repassivation” theory) is a modification of adsorption-displacement theory
of competitive adsorption between Cl- and O2- on the metal surface. In this theory of pit
initiation, breakdown potential (Ep) is referred a particular potential at which the
adsorption of aggressive anions on the metal surface displaces the adsorbed passivating
species.
Mechanical stresses at the weak sites or flaws on the oxide film resulting from the
electrostriction and surface tension can lead into the local breakdown of the passive film
[112]. However, in a non-aggressive environment (i.e., in the absence of chloride or other
Appendix A: Mechanisms and Models for Pit Initiation
401
aggressive anions) the passive film can reform very rapidly. By comparison, the
likelihood of the film reform will decrease rapidly in the presence of aggressive anions
and sufficiently high potential. As a result the metal surface at those defective sites
becomes activated and attacked by the formation of soluble non protective corrosion
products. In this model, the role of chloride ions is restricted to prevent repassivation
rather than promoting breakdown [18].
While considering this film breakdown mechanism, it should be remembered that
not all the breakdown events would result into pitting corrosion. According to this model,
breakdown will only lead to pitting corrosions where pit growth is possible [112]. It is
assumed that breakdown will always occur, but the passive film properties will influence
the rate of its occurrence.
Pit stability criteria play an important role in describing the pitting corrosion
mechanism using this model. According to this model, film breakdown occurs probably
with a different energy and at different spots on the film. Hence, there is a need to
establish some criteria that can describe the conditions needed pit to be able to grow.
Hence several researchers [114] suggested a critical factor for pit stability mainly
consisting of the product of pit depth and pit current density [112]. If the product (which
could be different for different metals/alloys) of this exceeds a certain value, pit is
supposed to be stable. So, it can be stated as summary that the film breaking model really
involves initiation based on pit growth stability [112].
Appendix A: Mechanisms and Models for Pit Initiation
402
A.4 Mechano - Chemical Model
This model/mechanism of pit initiation describes with the interaction between the
aggressive anions with the passive oxide film which could be under mechanically stressed
condition [18, 128, 133, 134, 328]. Mechanical stress in the anodic oxide film can arise
due to the reasons like: (i) interfacial tension, (ii) electrostriction pressure resulting from
the presence of a high electric field in the film, (iii) internal stress caused by the volume
ratio of the film and the metal, (iv) internal stress due to partial hydration or dehydration
of the film, or (iv) local stresses caused by the impurities. Under these operating stresses,
surface tension probably stabilizes the anodic oxide film. However, this surface tension
effect decreases with the increase in the film thickness and this will lead to the existence
of a critical film thickness above which breakdown of the passive film could occur.
Adsorption of the aggressive anions on the oxide film will lower the surface tension and
hence decrease the critical thickness for breakdown [133]. Thus breakdown potential at
the critical film thickness is found to depend on the anion concentration in the solution
[133]. Hoar [328] suggested that the adsorption of Cl- onto the film/solution interface
results in “peptization” due to the mutual repulsion of the adsorbed charged species.
When the repulsive forces are sufficiently high, the film cracks. Sato [133] suggested that
nature of oxide film breakdown depends on the mechanical properties of the film. Rigid
anhydrous metal oxide would result into brittle crack or fracture whereas visco-plasctic
hydrous oxide would result into plastic deformation or flow. Sato [133] suspected that in
case of plastic flow continuous breakdown can proceed by producing pores on top of the
anodic oxide film. However, any of these types of damages (i.e., either flaws or pores) in
the anodic oxide film would expose the underlying metal for dissolution.
Appendix A: Mechanisms and Models for Pit Initiation
403
A.5 Point Defect Model for Pit Initiation
Point defect model for pit initiation can also be correlated with the penetration
mechanism as described earlier [112]. In the point defect model, a passive film is
regarded as containing numerous point defects [18, 112, 154-156]. The major point
defects in an oxide film are assumed to be electron, and metal (cation) and oxygen (anion)
vacancies. Metal and oxide vacancies are in their equilibrium states at the metal/solution
and film/solution interface. Film thickening occurs when anion diffuses from the
film/solution interface to the metal/film interface. On the other hand cation diffusion
results only in dissolution.
Metal vacancies (or “metal holes”) are created at the metal/film interface due to the
diffusion of cation from the metal/film to the film/solution interfaces. These “metal
holes” will tend to submerge into the bulk of the metal and hence to “disappear” [155].
However, when the cation diffusion rate (i.e., the “metal hole” production rate) is higher
than the rate of “metal hole” submerge into the bulk, the metal holes start pilling up and
hence will form a void at the metal/film interface (this is the process of pit incubation).
When the void grow to a certain critical size, the passive film suffer local collapse and
thereby initiating pitting corrosion [18, 155].
Cl- ions are capable of being incorporated into passive film through the occupation
of oxygen vacancies by the Cl- ions. In order to main the charge neutrality, the number of
vacant cation sites in the neighbourhood must increase [156]. In this mechanism, anion
absorption increases the generation of cation vacancies across the barrier layer. Thus,
though the generation of cation vacancies is autocatalytic, but whether or not the film
breaks down depend on the relative rates with which the cation vacancies are transferred
across the barrier layer and are annihilated by emissions of cations from the metal into the
Appendix A: Mechanisms and Models for Pit Initiation
404
film. Passivity breakdown is though to occur at regions of the film that are characterized
by high cation vacancy diffusivities [156].
A.6 Localized Acidification Theory
Galvele [114] was the first to propose the “localized acidification mechanism”
where the induction time for pitting is described as the time required to achieve the
critical pH inside the pit initiation sites. It is also suggested that pit develops because of
the hydrolysis of corrosion products during transient cracks in the oxide film thus causing
acidification at the metal surface [134]. On the other hand, Smialowska [18] treated this
model as a “pit growth” model than “pit initiation” model. According to Smialowska [18]
pits can be formed only when the passive film if destroyed locally; however subsequent
pit growth will be determined by the formation of a pit solution that prevents the
repassivation.
Galvele [114] assumed that film breakdown occurs constantly even below the
pitting potential. In the presence of high enough electrode potential, a crack in the passive
film will promote metal dissolution followed by a hydrolysis reaction which will drop the
pH inside the pit. Galvele suggested that for each metal and alloy, a critical acidification
is necessary to render repassivation impossible and sustain pit activity. The pitting
potential does not have any special thermodynamic significance according to this model;
rather it is described as a potential necessary to reach the current density for the critical
x.i (i.e., pit depth.current density) [18].
Galvele [114] found that in most of the metals the critical pH is reached with x.i
values lower than 10-6 A/cm. Since at pit initiation conditions the current density the pit is
Appendix A: Mechanisms and Models for Pit Initiation
405
at least 1A/cm2, it is concluded that the necessary acidification can be obtained in pits as
small as 10-6 cm. This means that a crack in the passivating oxide film would give a
diffusion path long enough to reach the critical pH. If such cracks are present, it would
only be necessary to apply potential high enough to reach the above mentioned current
density.
Thus, Galvele’s model is based on the dissolution of a metal, cation hydrolysis,
diffusion and migration, which leads to the prediction of pit acidification.
A.7 Chemical Dissolution Theories
According to the theories aggressive anions and metal ions in oxide film form
complexes that cause localized thinning of the film leading to its breakdown [120, 135,
139, 155, 157]. The chemical dissolution theory suggests that small number of Cl- ions
jointly adsorbed around a cation in the film surface to form a high energy complex and
once formed, this complex will readily dissolve into the solution making the oxide film
thinner locally. The presence of a stronger anodic field at this site will rapidly transfer
another cation to the surface where it will meet more Cl-, complex with them, and thereby
enter the solution. Thus this process will become auto-catalytic and results in film
thinning [135, 155].
Nguyen and Foley [107, 139] have performed extensive study on the chemical
nature of aluminium corrosion. Nguyen experimental results support the idea that the first
steps in aluminium pitting in aggressive solutions involve adsorption of the anion on the
oxide film followed by chemical interaction to form a soluble species resulting in a
thinning of the film and direct contact with the metal surface.
Appendix A: Mechanisms and Models for Pit Initiation
406
According to Foley’s group the initiation of pitting of aluminium in halide solutions
proceeds in four consecutive steps:
i) Adsorption of the aggressive anion on the oxide film
ii) The chemical reaction of the adsorbed anion with the Al3+ in the oxide lattice:
ClOHAlClOlatticenHOinAlAl 22323 )().( →+ −+
iii) The thinning of the oxide film by dissolution
iv) The direct attack of the exposed metal by the aggressive anion with the formation of transient complexes which rapidly undergo hydrolysis:
−−+ →+ 43 4 AlClClAl
−+− ++→+ ClHClOHAlOHAlCl 32)(2 224
References
407
REFERENCES
1. E. Ghali, in Uhlig's Corrosion Handbook, R. W. Revie (Editor), John Wiley & Sons, p. 677, 2000.
2. J. G. Kaufman, in Handbook of Materials Selection, M. Kutz (Editor), John Wiley & Sons, Inc., New York, 2002.
3. N. P. Cheremisinoff, Materials Selection Deskbook, William Andrew Publishing, 1996.
4. E. A. Starke, Jr., and J. T. Staley, Progress in Aerospace Sciences, 32(2-3), 131-172, 1996.
5. I. J. Polmear, Light Alloys - Metallurgy of the Light Metals, Arnold, London, 1995.
6. P. Campestrini, E. P. M. van Westing, H. W. van Rooijen, and J. H. W. de Wit, Corrosion Science, 42(11), 1853-1861, 2000.
7. V. Guillaumin, and G. Mankowski, Corrosion Science, 41(3), 421-438, 1999. 8. G. S. Chen, M. Gao, and R. P. Wei, Corrosion, 52(1), 8-15, 1996. 9. R. G. Buchheit, R. P. Grant, P. F. Hlava, B. Mckenzie, and G. L. Zender, Journal
of the Electrochemical Society, 144(8), 2621-2628, 1997. 10. Y. Yoon, and R. G. Buchheit, Journal of the Electrochemical Society, 153(5),
151-155, 2006. 11. C. M. Liao, J. M. Olive, M. Gao, and R. P. Wei, Corrosion, 54(6), 451-458, 1998. 12. M. Shao, Y. Fu, R. Hu, and C. Lin, Materials Science and Engineering A, 344(1-
2), 323-327, 2003. 13. Z. Liu, P. H. Chong, A. N. Butt, P. Skeldon, and G. E. Thompson, Applied
Surface Science, 247(1-4), 294-299, 2005. 14. R. G. Buchheit, L. P. Montes, M. A. Martinez, J. Michael, and P. F. Hlava,
Journal of the Electrochemical Society, 146(12), 4424-4428, 1999. 15. J. F. Li, Z. Ziqiao, J. Na, and T. Chengyu, Materials Chemistry and Physics, 91(2-
3), 325-329, 2005. 16. T. Suter, and H. Bohni, Electrochimica Acta, 42(20-22), 3275-3280, 1997. 17. T. Suter, and H. Bohni, in Analytical Methods in Corrosion Science and
Engineering, P. Marcus and F. Mansfeld (Editors), Taylor & Francis Group, London, p. 649-696, 2006.
18. Z. Szklarska-Smialowska, Pitting and Crevice Corrosion, NACE International, Houston, Texas, 2005.
19. F. Moulinier, "Effects of a Surface Treatment on the Corrosion and Fatigue Behaviour of a High Strength Aluminium Alloy", University of Birmingham, PhD Thesis, 2003.
20. C. M. Rangell, M. H. Simplicio, and M. G. S. Ferreira, British Corrosion Journal, 27, 237, 1992.
21. E. M. Gutman, G. Solovioff, and D. Eliezer, Corrosion Science, 38(7), 1141-1145, 1996.
22. T. P. Hoar, and F. P. Ford, Journal of the Electrochemical Society, 120(8), 1013-1019, 1973.
23. F. Navai, and O. Debbouz, Journal of Materials Science, 34, 1073, 1999.
References
408
24. V. Vignal, C. Valot, R. Oltra, M. Verneau, and L. Condreusse, Corrosion Science, 44, 1477, 2002.
25. G. Bombara, Corrosion Science, 19(12), 991-999, 1979. 26. P. Peyre, X. Scherpereel, L. Berthe, C. Carboni, R. Fabbro, G. Beranger, and C.
Lemaitre, Materials Science and Engineering A, A280, 294, 2000. 27. F. Navai, Journal of Materials Science, 35, 5921-5925, 2000. 28. T. Suter, E. G. Webb, H. Bohni, and R. C. Alkire, Journal of the Electrochemical
Society, 148(5), B174-B185, 2001. 29. X. Liu, G. S. Frankel, B. Zoofan, and S. I. Rokhlin, Corrosion Science, 46(2),
405-425, 2004. 30. V. Vignal, R. Oltra, and C. Josse, Scripta Materialia, 49, 779-784, 2003. 31. R. Oltra, and V. Vignal, Corrosion Science, 49(1), 158-165, 2007. 32. V. Vignal, N. Mary, R. Oltra, and J. Peultier, Journal of the Electrochemical
Society, 153(9), B352-B357, 2006. 33. V. Vignal, N. Mary, C. Valot, R. Oltra, and L. Coudreuse, Electrochemical and
Solid-State Letters, 7(4), 39-42, 2004. 34. D. Chastell, P. Doig, P. E. J. Flewitt, and K. Ryan, Corrosion Science, 19(5), 335-
341, 1979. 35. Z. Zhao, and G. S. Frankel, "The Effects of Prior Deformation on Localized
Corrosion of Al Alloys," The Ohio State University Research Foundation, Columbus, OH, 2005.
36. B. J. Connolly, in Local Probe Techniques in Corrosion Research, R. Oltra (Editor), Woodhead Publishing Limited, Cambridge, UK, p. 155-166, 2007.
37. G. O. Ilevbare, O. Schneider, R. G. Kelly, and J. R. Scully, Journal of the Electrochemical Society, 151(8), 453-464, 2004.
38. O. Schneider, G. O. Ilevbare, J. R. Scully, and R. G. Kelly, Journal of the Electrochemical Society, 151(8), 465-472, 2004.
39. T. Suter, and R. C. Alkire, Critical Factors in Localized Corrosion III, Electrochemical Society, p. 118-129, 1998.
40. T. Suter, and R. C. Alkire, Journal of the Electrochemical Society, 148(1), B36-B42, 2001.
41. F. Andreatta, M. M. Lohrengel, H. Terryn, and J. H. W. de Wit, Electrochimica Acta, 48(20-22), 3239-3247, 2003.
42. N. Birbilis, and R. G. Buchheit, Journal of the Electrochemical Society, 152(4), B140-B151, 2005.
43. P. Schmutz, and G. S. Frankel, Journal of the Electrochemical Society, 145(7), 2295-2306, 1998.
44. P. Schmutz, and G. S. Frankel, Journal of the Electrochemical Society, 145(7), 2285-2295, 1998.
45. T. J. Warner, M. P. Schmidt, F. Sommer, and D. Bellot, Zeitschrift fur Metallkunde, 86(7), 494-501, 1995.
46. P. Leblanc, and G. S. Frankel, Journal of the Electrochemical Society, 149(6), 239-247, 2002.
47. R. M. Rynders, C. H. Paik, R. Ke, and R. C. Alkire, Journal of the Electrochemical Society, 141(6), 1439-1445, 1994.
48. J. O. Park, C. H. Paik, Y. H. Huang, and R. C. Alkire, Journal of the Electrochemical Society, 146(2), 517-523, 1999.
49. F. Hunkeler, and H. Bohni, Corrosion, 37(11), 645-650, 1981. 50. F. Hunkeler, and H. Bohni, Corrosion, 40(10), 534-540, 1984.
References
409
51. A. Sehgal, G. S. Frankel, B. Zoofan, and S. Rokhlin, Journal of the Electrochemical Society, 147(1), 140-148, 2000.
52. W. Zhang, and G. S. Frankel, Journal of the Electrochemical Society, 149(11), 510-519, 2002.
53. X. Liu, G. S. Frankel, B. Zoofan, and S. I. Rokhlin, Corrosion Science, 49(1), 139-148, 2007.
54. X. Liu, G. S. Frankel, B. Zoofan, and S. I. Rokhlin, Corrosion, 62(3), 217-230, 2006.
55. X. Zhao, G. S. Frankel, B. Zoofan, and S. I. Rokhlin, Corrosion, 59(11), 1012-1018, 2003.
56. ASM Speciality Handbook: Aluminum and Aluminum Alloys, ASM International, Ohio, 1993.
57. C. R. Brooks, Heat Treatment, Structure and Properties of Nonferrous Alloys, American Society for Metals, Ohio, 1990.
58. J. B. Dwight, Aluminum Design and Construction, Spon Press, UK, 1999. 59. Smithells Metals Reference Book, Butterworth Heinemann, Oxford, 2004. 60. G. E. Dieter, Mechanical Metallurgy, McGraw-Hill, New York, 1986. 61. J. T. Staley, in Aluminum Alloys - Contemporary Reserach and Applications, A.
K. Vasudevan and R. D. Doherty (Editors), Academic Press, Inc., New York, 1989.
62. K. R. Van Horn, Aluminum, Vol. I: Properties, Physical Metallurgy, and Phase Diagrams, ASM, Metal Park, Ohio, 1967.
63. L. F. Mondolfo, Aluminium Alloys: Structures and Properties, Butterworth & Co, London, 1976.
64. Properties of Wrought Aluminium and Aluminium Alloys, Metals Handbook, Properties and Selection: Nonferrous Alloys and Special Purpose Materials, vol. 2, ASM International, Material Park, Ohio, 1990.
65. ASM Handbook, Heat Treating, vol. 4, ASM International, Material Park, Ohio, 1991.
66. M. Nakai, and T. Eto, Materials Science and Engineering A, 285(1-2), 62-68, 2000.
67. G. T. Hahn, and A. R. Rosenfield, Metallurgical Transactions A (Physical Metallurgy and Materials Science), 6A(4), 653-668, 1975.
68. S. Suresh, Fatigue of Materials, Cambridge University Press, Cambridge, 1998. 69. D. Sigler, M. C. Montpetit, and W. L. Haworth, Metallurgical Transactions A
(Physical Metallurgy and Materials Science), 14A(5), 931-938, 1983. 70. M. Jariyaboon, "Corrosion of Friction Stir Welds in High Strength Aluminium
Alloys", University of Birmingham, PhD Thesis, 2006. 71. MatWeb - Material Property Data, 2007: http://www.matweb.com. 72. R. H. Brown, and L. A. Willey, in Aluminium : Properties, Physical Metallurgy
and Phase Diagrams, K. R. V. Horn (Editor), Americal Society for Metals, Ohio, p. 53, 1967.
73. P. R. Sperry, Transactions of the ASM, 48, 904-918, 1956. 74. G. O. Ilevbare, and J. R. Scully, Journal of the Electrochemical Society, 148(5),
196-207, 2001. 75. M. Gao, C. R. Feng, and R. P. Wei, Metallurgical and Materials Transactions A:
Physical Metallurgy and Materials Science, 29A(4), 1145-1151, 1998. 76. A. E. Hughes, R. J. Taylor, K. J. H. Nelson, B. R. W. Hinton, and L. Wilson,
Materials Science and Technology, 12, 928-936, 1996.
References
410
77. G. O. Ilevbare, and J. R. Scully, Corrosion, 57(2), 134-152, 2001. 78. M. Qian, J. A. Taylor, J. Y. Yao, M. J. Couper, and D. H. StJohn, Journal of Light
Metals, 1, 187-193, 2001. 79. F. King, Aluminium and Its Alloys, John Wiley and Sons, New York, 1987. 80. H. M. Obispo, L. E. Murr, R. M. Arrowood, and E. A. Trillo, Journal of Materials
Science, 35, 3479-3495, 2000. 81. E. A. DeBartolo, and B. M. Hillberry, International Journal of Fatigue, 20(10),
727-735, 1998. 82. R. G. Buchheit, R. K. Boger, and M. W. Donohue, Symposium of the Seawater
Corrosion, The Electrochemical Society, p. 205, 1999. 83. V. Radmilovic, R. Kilaas, U. Dahmen, and G. J. Shiflet, Acta Materialia, 47(15-
16), 3987-3997, 1999. 84. S. Ghosh, A. J. Davenport, and B. J. Connolly, "Effect of Surface Treatment and
S-Phase Removal on the Corrosion Properties of AA2024-T3", (Unpublished Work), University of Birmingham, 2007.
85. L. M. Wang, H. M. Flower, and T. C. Lindley, Scripta Materialia, 41(4), 391-396, 1999.
86. M. E. Fine, Metallurgical Transactions A: Physical Metallurgy and Materials Science, 6A(4), 625-630, 1975.
87. M. Rosen, E. Horowitz, L. Swartzendruber, S. Fick, and R. Mehrabian, Materials Science and Engineering, 53(2), 191-198, 1982.
88. Y. Gefen, M. Rosen, and A. Rosen, Material Science and Engineering, 8(3), 181-188, 1971.
89. N. Sen, and D. R. F. West, Journal of the Institute of Metals, 97, 87-92, 1969. 90. H.-C. Shih, N.-J. Ho, and J. C. Huang, Metallurgical and Materials Transactions
A: Physical Metallurgy and Materials Science, 27A(9), 2479-2494, 1996. 91. A. K. Gupta, P. Gaunt, and M. C. Chaturvedi, Philosophical Magazine A: Physics
of Condensed Matter: Structure, Defects and Mechanical Properties, 55(3), 375-387, 1987.
92. A. K. Jena, A. K. Gupta, and M. C. Chaturvedi, Acta Metallurgica, 37(3), 885-895, 1989.
93. V. Radmilovic, G. Thomas, G. J. Shiflet, and E. A. Starke, Jr., Scripta Metallurgica, 23(7), 1141-1146, 1989.
94. S. P. Ringer, T. Sakurai, and I. J. Polmear, Acta Materialia, 45(9), 3731-3744, 1997.
95. W. Zhang, and G. S. Frankel, Electrochimica Acta, 48(9), 1193-1210, 2003. 96. in Metals Handbook: Corrosion, E. H. Hollingsworth and H. Y. Hunsicker
(Editors), ASM International, Material Park, OH, USA, 1987. 97. J. R. Davis, Corrosion of Aluminium and Aluminium Alloys, ASM International,
Materials Park, OH, USA, 1991. 98. M. Pourbaix, Atlas of Electrochemical Equilibria in Aqueous Solutions, National
Association of Corrosion Engineers, Houston, Texas, 1974. 99. P. Hulser, U. A. Kruger, and F. Beck, Corrosion Science, 38(1), 47-57, 1996. 100. K. C. Emregul, and A. A. Aksut, Corrosion Science, 42(12), 2051-2067, 2000. 101. R. D. Armstrong, and V. J. Braham, Corrosion Science, 38(9), 1463-1471, 1996. 102. S. M. Moon, and S. I. Pyun, Corrosion Science, 39(2), 399-408, 1997. 103. K. Nisancioglu, The 3rd International Conference on Aluminium Alloys,
Trondheim, Norway, p. 239-259, 1992. 104. E. J. Lee, and S. I. Pyun, Corrosion Science, 37(1), 157-168, 1995.
References
411
105. Z. Szklarska-Smialowska, Corrosion Science, 41, 1743-1767, 1999. 106. H. Bohni, in Uhlig's Corrosion Handbook, R. W. Revie (Editor), John Wiley &
Sons, p. 173-190, 2000. 107. T. H. Nguyen, and R. T. Foley, Journal of the Electrochemical Society, 126(11),
1855-1860, 1979. 108. K. Srinivasa Rao, and K. Prasad Rao, Transactions of the Indian Institute of
Metals, 57(6), 593-610, 2004. 109. T. Okada, Corrosion Science, 31, 453-458, 1990. 110. R. T. Foley, Corrosion (Houston, TX, United States), 42(5), 277-288, 1986. 111. M. G. Fontana, Corrosion Engineering, McGraw-Hill Book Company, Singapore,
1987. 112. G. S. Frankel, Journal of the Electrochemical Society, 145(6), 2186-2198, 1998. 113. J. R. Galvele, Critical Factors in Localized Corrosion, The Electrochemical
Society, Inc, p. 94-105, 1992. 114. J. R. Galvele, Journal of the Electrochemical Society, 123(4), 464-474, 1976. 115. P. L. Cabot, F. A. Centellas, J. A. Garrido, E. Perez, and H. Vidal, Electrochimica
Acta, 36(1), 179-187, 1991. 116. T. R. Beck, Electrochimica Acta, 33(10), 1321-1327, 1988. 117. Z. Szklarska-Smialowska, Corrosion Science, 33(8), 1193-1202, 1992. 118. L. Tomcsanyi, K. Varga, I. Bartik, G. Horanyi, and E. Maleczki, Electrochimica
Acta, 34(6), 855-859, 1989. 119. R. Ambat, and E. S. Dwarakadasa, Journal of Applied Electrochemistry, 24(9),
911-916, 1994. 120. E. McCafferty, Corrosion Science, 45(7), 1421-1438, 2003. 121. Z. Szklarska-Smialowska, Pitting Corrosion of Metals, NACE, Houston, TX,
1986. 122. K. Kowal, J. DeLuccia, J. Y. Josefowicz, C. Laird, and G. C. Farrington, Journal
of the Electrochemical Society, 143(8), 2471-2481, 1996. 123. R. G. Buchheit, R. K. Boger, M. C. Carroll, R. M. Leard, C. Paglia, and J. L.
Searles, JOM, 53(7), 29-33, 2001. 124. G. T. Burstein, L. L. Shreir, and R. A. Jarman, Corrosion, Volume 1, Elsevier
Butterworth-Heinemann, 1994. 125. I. T. E. Fonseca, N. Lima, J. A. Rodrigues, M. I. S. Pereira, J. C. S. Salvador, and
M. G. S. Ferreria, Electrochemistry Communications, 4(5), 353-357, 2002. 126. D. E. Williams, J. Stewart, and P. H. Balkwill, Critical Factors in Localized
Corrosion, The Electrochemical Society Softbound Proceeding Series, NJ, p. 36, 1992.
127. G. S. Frankel, L. Stockert, F. Hunkeler, and H. Bohni, Corrosion, 43, 429, 1987. 128. J. Kruger, and K. Rhyne, Nuclear and Chemical Waste Management, 3, 205,
1982. 129. H. Bohni, and H. H. Uhlig, Journal of the Electrochemical Society, 116, 906,
1969. 130. H. P. Leckie, and H. H. Uhlig, Journal of the Electrochemical Society, 113, 1262,
1966. 131. J. A. Richardson, and G. C. Wood, Journal of the Electrochemical Society, 10,
313, 1970. 132. M. A. Heine, and M. J. Pryor, Journal of the Electrochemical Society, 114, 1001,
1967. 133. N. Sato, Electrochemica Acta, 16, 1683-1692, 1971.
References
412
134. S. T. Pride, "Metastable pitting of aluminum and Al-Cu alloys", The University of Virginia, PhD Thesis, 1997.
135. T. P. Hoar, and W. R. Jackob, Nature, 216, 1299-1301, 1967. 136. H. H. Strehblow, in Corrosion Mechanisms in Theory and Practice, P. Marcus
and J. Oudar (Editors), Marcel Dekker, New York, 1995. 137. J. A. Richardson, and G. C. Wood, Corrosion Science, 10(5), 313-323, 1970. 138. R. T. Foley, and T. H. Nguyen, Journal of the Electrochemical Society, 129(3),
464-467, 1982. 139. T. H. Nguyen, and R. T. Foley, Journal of the Electrochemical Society, 129(1),
27-32, 1982. 140. M. F. Abd Rabbo, J. A. Richardson, and G. C. Wood, Corrosion Science, 18(2),
117-123, 1978. 141. J. O. Bockris, and L. V. Minevski, Journal of Electroanalytical Chemistry, 349(1-
2), 375-414, 1993. 142. J. O. M. Bockris, and Y. Kang, Journal of Solid State Electrochemistry, 1(1), 17-
35, 1997. 143. P. M. Natishan, W. E. O'Grady, E. McCafferty, D. E. Ramaker, K. Pandya, and A.
Russell, Journal of the Electrochemical Society, 146(5), 1737-1740, 1999. 144. S. Y. Yu, W. E. O'Grady, D. E. Ramaker, and P. M. Natishan, Journal of the
Electrochemical Society, 147(8), 2952-2958, 2000. 145. A. Kolics, A. S. Besing, P. Baradlai, R. Haasch, and A. Wieckowski, Journal of
the Electrochemical Society, 148(7), 251-259, 2001. 146. M. Koudelkova, J. Augustynski, and H. Berthou, Journal of the Electrochemical
Society, 124(8), 1165-1168, 1977. 147. M. J. Pryor, Corrosion Science, 11(6), 463-464, 1971. 148. E. J. Lee, and S. I. Pyun, Corrosion Science, 37(1), 157-168, 1995. 149. J. A. Richardson, and G. C. Wood, Journal of the Electrochemical Society,
120(2), 193-202, 1973. 150. B. N. Stirrup, N. A. Hampson, and I. S. Midgley, Journal of Applied
Electrochemistry, 5(3), 229-235, 1975. 151. Z. A. Foroulis, and M. J. Thubrikar, Journal of the Electrochemical Society,
122(10), 1296-1301, 1975. 152. J. R. Galvele, and S. M. d. De Micheli, Corrosion Science, 10(11), 795-807, 1970. 153. E. McCafferty, Corrosion Science, 37(3), 481-492, 1995. 154. C. Y. Chao, L. F. Lin, and D. D. Macdonald, Journal of the Electrochemical
Society, 128(6), 1187-1194, 1981. 155. L. F. Lin, C. Y. Chao, and D. D. Macdonald, Journal of the Electrochemical
Society, 128(6), 1194-1198, 1981. 156. D. D. Macdonald, Journal of the Electrochemical Society, 139(12), 3434-3449,
1992. 157. T. H. Nguyen, and R. T. Foley, Journal of the Electrochemical Society, 127(12),
2563-2566, 1980. 158. A. R. Trueman, Corrosion Science, 47(9), 2240-2256, 2005. 159. F. Sato, and R. C. Newman, Corrosion, 54(12), 955-963, 1998. 160. T. T. Lunt, S. T. Pride, J. R. Scully, J. L. Hudson, and A. S. Mikhailov, Journal of
the Electrochemical Society, 144(5), 1620-1629, 1997. 161. B. Wu, J. R. Scully, J. L. Hudson, and A. S. Mikhailov, Journal of the
Electrochemical Society, 144(5), 1614-1620, 1997.
References
413
162. G. S. Frankel, L. Stockert, F. Hunkeler, and H. Boehni, Corrosion, 43(7), 429-436, 1987.
163. P. C. Pistorius, and G. T. Burnstein, Corrosion Science, 33(12), 1885-1897, 1992. 164. P. C. Pistorius, and G. T. Burnstein, Corrosion Science, 36(3), 525-538, 1994. 165. D. E. Williams, J. Stewart, and P. H. Balkwill, Corrosion Science, 36(7), 1213-
1235, 1994. 166. D. W. Buzza, and R. C. Alkire, Journal of the Electrochemical Society, 142(4),
1104-1111, 1995. 167. S. T. Pride, J. R. Scully, and J. L. Hudson, Journal of the Electrochemical Society,
141(11), 3028-3040, 1994. 168. I. L. Muller, and J. R. Galvele, 17(3), 179-193, 1977. 169. A. J. Davenport, "Lecture Notes - Mechanisms of Corrosion of Metals in Aqueous
Environment ", 2005. 170. R. P. Wei, C.-M. Liao, and M. Gao, Metallurgical and Materials Transactions A:
Physical Metallurgy and Materials Science, 29A(4), 1153-1160, 1998. 171. T. J. R. Leclere, and R. C. Newman, Journal of the Electrochemical Society,
149(2), B52-B56, 2002. 172. M. B. Vukmirovic, N. Vasiljevic, N. Dimitrov, and K. Sieradzki, Journal of the
Electrochemical Society, 150(1), B10-B15, 2003. 173. N. Dimitrov, J. A. Mann, and K. Sieradzki, Journal of the Electrochemical
Society, 146(1), 98-102, 1999. 174. M. A. Alodan, and W. H. Smyrl, Journal of the Electrochemical Society, 144(10),
L282-L284, 1997. 175. M. B. Vukmirovic, N. Dimitrov, and K. Sieradzki, Journal of the Electrochemical
Society, 149(9), 428-439, 2002. 176. D. A. Jones, Principles and Prevention of Corrosion, Macmillan Publishing
Company, NY, USA, 1992. 177. H. H. Uhlig, Corrosion Science, 19(11), 777-791, 1979. 178. H. H. Uhlig, and R. W. Revie, Corrosion and Corrosion Control, John Wiley &
Sons, NY, USA, 1991. 179. C. M. Liao, and R. P. Wei, Electrochimica Acta, 45(6), 881-888, 1999. 180. C. Blanc, and G. Mankowski, Corrosion Science, 40(2-3), 411-429, 1998. 181. R. G. Buchheit, Materials Science Forum, 331 (II), 1641-1646, 2000. 182. R. G. Buchheit, M. A. Martinez, and L. P. Montes, Journal of the Electrochemical
Society, 147(1), 119-124, 2000. 183. N. Missert, J. C. Barbour, R. G. Copeland, and J. E. Mikkalson, JOM, 53(7), 34-
36, 2001. 184. G. Chen, M. Williams, J. Malmberg, J. S. Kim, and D. A. Buttry, Corrosion and
Corrosion Prevention of Low Density Metals and Alloys - Proceedings of the International Symposium, The Electrochemical Society, p. 29-40, 2000.
185. N. Missert, R. G. Copeland, Y. Kim, and R. Buchheit, Corrosion and Protection of Light Metal Alloys, The Electrochemical Soceity, p. 55-61, 2003.
186. N. Dimitrov, J. A. Mann, M. Vukmirovic, and K. Sieradzki, Journal of the Electrochemical Society, 147(9), 3283-3285, 2000.
187. J. C. Seegmiller, R. C. Bazito, and D. A. Buttry, Electrochemical and Solid-State Letters, 7(1), 1-4, 2004.
188. M. Kendig, S. Jeanjaquet, R. Buchheit, H. Guan, and R. M. Leard, Corrosion and Corrosion Prevention of Low Density Metals and Alloys, The Electrochemical Society, p. 1-9, 2001.
References
414
189. A. Kolics, A. S. Besing, T. Suter, H. Bohni, C. H. Paik, R. C. Alkire, and A. Wieckowski, Corrosion and Corrosion Prevention of Low Density Metals and Alloys, The Electrochemical Society, p. 10-18, 2000.
190. R. K. Boger, and R. Buchheit, Corrosion and Corrosion Prevention of Low Density Metals and Alloys, The Electrochemical Society, p. 19-28, 2000.
191. R. G. Buchheit, Journal of the Electrochemical Society, 142(11), 3994-3996, 1995.
192. Z. Szklarska-Smialowska, Corrosion, 27(6), 223-233, 1971. 193. A. Kolics, A. S. Besing, T. Suter, H. Bohni, C. H. Paik, R. C. Alkire, and A.
Wieckowski, Corrosion and Corrosion Prevention of Low Density Metals and Alloys, The Electrochemical Society, p. 10-18, 2001.
194. R. K. Boger, and R. Buchheit, Corrosion and Corrosion Prevention of Low Density Metals and Alloys, The Electrochemical Society, p. 19-28, 2001.
195. M. Buchler, J. Kerimo, F. Guillaume, and W. H. Smyrl, Journal of the Electrochemical Society, 147(10), 3691-3699, 2000.
196. M. Buchler, T. Watari, and W. H. Smyrl, Corrosion Science, 42(9), 1661-1668, 2000.
197. M. A. Jakab, D. A. Little, and J. R. Scully, Journal of the Electrochemical Society, 152(8), 311-320, 2005.
198. M. A. Alodan, and W. H. Smyrl, Journal of the Electrochemical Society, 145(5), 1571-1577, 1998.
199. N. Missert, R. G. Copeland, J. C. Barbour, J. E. Mikkalson, and H. Isaacs, Proceedings - Electrochemical Society, 2000-23(Corrosion and Corrosion Prevention of Low Density Metals and Alloys), 239-246, 2001.
200. N. Missert, R. G. Copeland, Y. Kim, and R. Buchheit, Corrosion and Protection of Light Metal Alloys - Proceedings of the International Symposium, Oct 12-17 2003, Orlando, FL., United States, Electrochemical Society Inc., Pennington, United States, p. 55-61, 2003.
201. B. Mazurkiewicz, Corrosion Science, 23(7), 687-691, 1983. 202. B. Mazurkiewicz, and A. Piotrowski, Corrosion Science, 23(7), 697-701, 1983. 203. K. A. Yasakau, M. L. Zheludkevich, S. V. Lamaka, and M. G. S. Ferreira, Journal
of Physical Chemistry B, 110(11), 5515-5528, 2006. 204. G. O. Ilevbare, and J. R. Scully, Corrosion (Houston, TX, United States), 57(2),
134-152, 2001. 205. K. Sugimoto, K. Hoshino, M. Kageyama, S. Kageyama, and Y. Sawada,
Corrosion Science, 15(11-12), 709-720, 1975. 206. M. J. Pryor, and J. C. Fister, Journal of the Electrochemical Society, 131(6), 1230-
1235, 1984. 207. S. J. Ketcham, and F. H. Haynie, Corrosion, 19(7), 242-246, 1963. 208. K. Nisancioglu, K. Y. Davanger, O. Strandmyr, and H. Holton, Journal of the
Electrochemical Society, 128(7), 1523-1526, 1981. 209. Y. Baek, and G. S. Frankel, Journal of the Electrochemical Society, 150(1), B1-
B9, 2003. 210. C. Blanc, B. Lavelle, and G. Mankowski, Corrosion Science, 39(3), 495-510,
1997. 211. D. Zhu, and W. J. van Ooij, Corrosion Science, 45(10), 2163-2175, 2003. 212. P. Campestrini, H. Terryn, A. Hovestad, and J. H. W. de Wit, Surface &
Coatings Technology, 176(3), 365-381, 2004.
References
415
213. M. L. Zheludkevich, K. A. Yasakau, S. K. Poznyak, and M. G. S. Ferreira, Corrosion Science, 47(12), 3368-3383, 2005.
214. J. H. W. De Wit, Electrochimica Acta, 46(24-25), 3641-3650, 2001. 215. A. Kolics, A. S. Besing, P. Baradlai, and A. Wieckowski, Journal of the
Electrochemical Society, 150(11), 512-516, 2003. 216. A. J. Aldykewicz, Jr., H. S. Isaacs, and A. J. Davenport, Journal of the
Electrochemical Society, 142(10), 3342-3350, 1995. 217. J. R. Scully, in Encyclopedia of Electrochemistry: Corrosion and Oxide Films, A.
J. Bard and M. Stratmann (Editors), WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, p. 344-384, 2003.
218. A. Rota, and H. Boehni, Materials Science Forum, 44-45(Electrochem. Methods Corros. Res.), 177-190, 1989.
219. A. Rota, and H. Boehni, Werkstoffe und Korrosion, 40(4), 219-228, 1989. 220. A. Garner, and D. Tromans, Corrosion, 35(2), 55-60, 1979. 221. B. W. Lifka, in Corrosion Engineering Handbook, P. A. Schweitzer (Editor),
Marcel Dekker, Inc., NY, p. 99-155, 1996. 222. S. Maitra, and G. C. English, Metallurgical Transactions A (Physical Metallurgy
and Materials Science), 12A(3), 535-541, 1981. 223. F. Andreatta, H. Terryn, and J. H. W. De Wit, Electrochimica Acta, 49(17-18),
2851-2862, 2004. 224. P. Doig, P. E. J. Flewitt, and J. W. Edington, Corrosion, 33(6), 217-221, 1977. 225. P. Doig, and J. W. Edington, Nuclear Technology, 461-476, 1975. 226. V. S. Sinyavskii, V. V. Ulanova, and V. D. Kalinin, Protection of Metals
(Translation of Zashchita Metallov), 40(5), 481-490, 2004. 227. T. Ramgopal, P. I. Gouma, and G. S. Frankel, Corrosion, 58(8), 687-697, 2002. 228. K. Urushino, and K. Sugimoto, Corrosion Science, 19(4), 225-236, 1979. 229. M. Yasuda, F. Weinberg, and D. Tromans, Journal of the Electrochemical
Society, 137(12), 3716-3723, 1990. 230. B. Venkataraman, K. G. Rao, and M. Saletore, Science & Technology of Al-Li
Alloys, HAL, Bangalore, p. 209-212, 1989. 231. Q. Zhang, F. B. Song, D. Li, and X. Y. Ding, Materials Science Forum, 396-
402(Pt. 3, Aluminium Alloys 2002), 1479-1484, 2002. 232. D. A. Little, B. J. Connolly, and J. R. Scully, Corrosion Science, 49(2), 347-372,
2007. 233. W. Zhang, and G. S. Frankel, Electrochemical and Solid-State Letters, 3(6), 268-
270, 2000. 234. W. Zhang, S. Ruan, D. A. Wolfe, and G. S. Frankel, Corrosion Science, 45(2),
353-370, 2003. 235. S. J. Fox, "An In Situ X-ray Tomography Study of the Growth and Development of
Localized Corrosion in Friction Stir Welded AA2024-T351", University of Birmingham, MRes Thesis, 2007.
236. G. Kotsikos, J. M. Sutcliffe, and N. J. H. Holroyd, Corrosion Science, 42(1), 17-33, 2000.
237. E. M. Gutman, Mechanochemistry of Solid Surfaces, World Scientific, Singapore, 1994.
238. E. M. Gutman, Mechanochemistry of Materials, Cambridge International Science Publishing, Cambridge, 1998.
239. P. L. Harrison, Corrosion Science, 7(12), 789-794, 1967.
References
416
240. V. Vignal, R. Oltra, and N. Mary, Paper No. 04446, NACE, New Orleans, US, p., 2004.
241. X. Liu, and G. S. Frankel, Corrosion Science, 48(10), 3309-3329, 2006. 242. V. S. Sinyavskii, and A. M. Semenov, Protection of Metals (Translation of
Zashchita Metallov), 38(2), 132-140, 2002. 243. O. Lunder, B. Olsen, and K. Nisancioglu, International Journal of Adhesion and
Adhesives, 22(2), 143-150, 2002. 244. L. Kozma, and I. Olefjord, Materials Science and Technology, 3(10), 860-874,
1987. 245. P. Campestrini, E. P. M. Van Westing, and J. H. W. De Wit, Electrochimica Acta,
46(16), 2553-2571, 2001. 246. A. Afseth, J. H. Nordlien, G. M. Scamans, and K. Nisancioglu, Corrosion
Science, 44(11), 2543-2559, 2002. 247. A. Afseth, J. H. Nordlien, G. M. Scamans, and K. Nisancioglu, Corrosion
Science, 44(11), 2491-2506, 2002. 248. K. Nisancioglu, J. H. Nordlien, A. Afseth, and G. M. Scamans, Charlottesville,
VA, USA, Trans Tech Publications, p. 111-126, 2000. 249. O. Lunder, K. F. Heen, and K. Nisancioglu, Corrosion, 60(7), 622-631, 2004. 250. G. S. Frankel, and R. L. McCreery, Electrochemical Society Interface, 10(4), 34-
38, 2001. 251. K. Nisancioglu, O. Lunder, and H. Holtan, Corrosion, 41(5), 247-257, 1985. 252. O. Lunder, and K. Nisancioglu, Corrosion, 44(7), 414-422, 1988. 253. C. E. Moffitt, D. M. Wieliczka, and H. K. Yasuda, Surface and Coating
Technology, 137, 188-196, 2001. 254. R. Pinner, and S. Wernick, The Surface Treatment and Finishing of Aluminium
and its Alloys, Robert Draper Ltd., Teddington, 1972. 255. E. V. Koroleva, G. E. Thompson, G. Hollrigl, and M. Bloeck, Corrosion Science,
41(8), 1475-1495, 1999. 256. A. E. Hughes, G. Theodossiou, S. Elliott, T. G. Harvey, P. R. Miller, J. D.
Gorman, and J. K. Paterson, Materials Science and Technology, 17, 1642-1652, 2001.
257. A. J. Davenport, and B. Liu, Corrosion and Corrosion Prevention of Low Density Metals and Alloys, The Electrochemical Society, p. 41-46, 2001.
258. A. J. Davenport, F. A. M. Moulinier, B. Liu, and P. C. Morgan, Aluminium Surface Science and Technology, Manchester, UK, p., 2000.
259. K. J. H. Nelson, A. E. Hughes, R. J. Taylor, B. R. W. Hinton, L. Wilson, and M. Henderson, Materials Science and Technology, 17, 1211-1221, 2001.
260. A. E. Hughes, K. J. H. Nelson, and P. R. Miller, Materials Science and Technology, 15, 1124-1132, 1999.
261. E. Maire, J. Y. Buffiere, L. Salvo, J. J. Blandin, W. Ludwig, and J. M. Letang, Advanced Engineering Materials, 3(8), 539-546, 2001.
262. M. Stampanoni, G. Borchert, P. Wyss, R. Abela, B. Patterson, S. Hunt, D. Vermeulen, and P. Ruegsegger, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 491(1-2), 291-301, 2002.
263. M. Stampanoni, R. Abela, G. Borchert, and B. D. Patterson, Developments in X-Ray Tomography IV, Aug 4-6 2004, Denver, CO, United States, International Society for Optical Engineering, Bellingham, WA 98227-0010, United States, p. 169-181, 2004.
References
417
264. B. D. Patterson, R. Abela, H. Auderset, Q. Chen, F. Fauth, F. Gozzo, G. Ingold, H. Kuhne, M. Lange, D. Maden, D. Meister, P. Pattison, T. Schmidt, B. Schmitt, C. Schulze-Briese, M. Shi, M. Stampanoni, and P. R. Willmott, Nuclear Instruments and Methods in Physics Research, Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, 540(1), 42-67, 2005.
265. M. Stampanoni, P. Wyss, R. Abela, G. Borchert, D. Vermeulen, and P. Ruegsegger, Developments in X-Ray Tomography III, Aug 2-3 2001, San Diego, CA, United States, The International Society for Optical Engineering, p. 42-53, 2001.
266. L. Salvo, P. Cloetens, E. Maire, S. Zabler, J. J. Blandin, J. Y. Buffiere, W. Ludwig, E. Boller, D. Bellet, and C. Josserond, Nuclear Instruments and Methods in Physics Research, Section B: Beam Interactions with Materials and Atoms, 200(SUPPL), 273-286, 2003.
267. S. R. Stock, International Materials Reviews, 44(4), 141-164, 1999. 268. E. Maire, J. Y. Buffiere, R. Mokso, P. Cloetens, and W. Ludwig, Materials
Science Forum, 519-521, 1367-1372, 2006. 269. G. N. Hounsfield, British Journal of Radiology, 46(552), 1016-1022, 1973. 270. G. Peix, N. Duvauchelle, and N. Freud, in X-Ray Tomography in Material
Science, J. Baruchel, J. Y. Buffiere, E. Maire, P. Merle and G. Peix (Editors), Hermes Science, Paris, p. 15-27, 2000.
271. C. R. Coutant, and A. Marc, in X-Ray Tomography in Material Science, J. Baruchel, J. Y. Buffiere, E. Maire, P. Merle and G. Peix (Editors), Hermes Science, Paris, p. 61-75, 2000.
272. L. Salvo, P. Cloetens, E. Maire, S. Zabler, J. J. Bladin, J. Y. Buffiere, W. Ludwig, E. Boller, D. Bellet, and C. Josserond, Nuclear Instruments & Methods in Physics Research B, 200, 273-286, 2003.
273. J. Baruchel, E. Boller, P. Cloetens, W. Ludwig, and F. Peyrin, in X-Ray Tomography in Material Science, J. Baruchel, J. Y. Buffiere, E. Maire, P. Merle and G. Peix (Editors), Hermes Science, Paris, p. 29-43, 2000.
274. L. Babout, W. Ludwig, E. Maire, and J. Y. Buffiere, Nuclear Instruments and Methods in Physics Research, Section B: Beam Interactions with Materials and Atoms, 200(SUPPL), 303-307, 2003.
275. J.-Y. Buffiere, E. Maire, P. Cloetens, G. Lormand, and R. Fougeres, Acta Materialia, 47(5), 1613-1625, 1999.
276. L. Babout, E. Maire, J. Y. Buffiere, and R. Fougeres, Acta Materialia, 49(11), 2055-2063, 2001.
277. E. Maire, J. C. Grenier, and L. Babout, Materials Science Forum, 519-521, 821-827, 2006.
278. E. Maire, and J. Y. Buffiere, in X-Ray Tomography in Material Science, J. Baruchel, J. Y. Buffiere, E. Maire, P. Merle and G. Peix (Editors), Hermes Science Publications, Paris, p. 115-125, 2000.
279. E. Maire, L. Babout, J. Y. Buffiere, and R. Fougeres, Materials Science & Engineering A (Structural Materials: Properties, Microstructure and Processing), A319-321, 216-219, 2001.
280. D. Bernard, G. L. Vignoles, and J. M. Heintz, in X-Ray Tomography in Material Science, J. Baruchel, J. Y. Buffiere, E. Maire, P. Merle and G. Peix (Editors), Hermes Science Publications, Paris, p. 177-191, 2000.
281. T. J. Marrow, J.-Y. Buffiere, P. J. Withers, G. Johnson, and D. Engelberg, International Journal of Fatigue, 26(7), 717-725, 2004.
References
418
282. T. J. Marrow, H. Cetinel, M. Al-Zalmah, S. Macdonald, P. J. Withers, and J. Walton, Fatigue and Fracture of Engineering Materials and Structures, 25(7), 635-648, 2002.
283. K. H. Khor, J.-Y. Buffiere, W. Ludwig, H. Toda, H. S. Ubhi, P. J. Gregson, and I. Sinclair, Journal of Physics Condensed Matter: Synchroton Radiation for Advanced Materials Analysis and Processing, 16(33), 3511-3515, 2004.
284. H. Toda, I. Sinclair, J.-Y. Buffiere, E. Maire, T. Connolley, M. Joyce, K. H. Khor, and P. Gregson, Philosophical Magazine, 83(21), 2429-2448, 2003.
285. A. Guvenilir, T. M. Breunig, J. H. Kinney, and S. R. Stock, Acta Materialia, 45(5), 1977-1987, 1997.
286. A. Guvenilir, and S. R. Stock, Fatigue and Fracture of Engineering Materials & Structures, 21(4), 439-450, 1998.
287. T. J. Marrow, L. Babout, A. P. Jivkov, P. Wood, D. Engelberg, N. Stevens, P. J. Withers, and R. C. Newman, Journal of Nuclear Materials, 352(1-3), 62-74, 2006.
288. B. J. Connolly, D. A. Homer, S. J. Fox, A. J. Davenport, C. Padovani, S. Zhou, A. Turnbull, M. Preuss, N. P. Stevens, T. J. Marrow, J. Y. Buffiere, E. Boller, A. Groso, and M. Stampanoni, Materials Science and Technology, 22(9), 1076-1085, 2006.
289. A. J. Davenport, C. Padovani, B. J. Connolly, N. P. C. Stevens, T. A. W. Beale, A. Groso, and M. Stampanoni, Electrochemical and Solid-State Letters, 10(2), 5-8, 2007.
290. L. Babout, T. J. Marrow, D. Engelberg, and P. J. Withers, Materials Science and Technology, 22(9), 1068-1075, 2006.
291. G. Y. Baaklini, R. T. Bhatt, A. J. Eckel, P. Engler, M. G. Castelli, and R. W. Rauser, Materials Evaluation, 53(9), 1040-1044, 1995.
292. J. H. Kinney, S. R. Stock, M. C. Nichols, U. Bonse, T. M. Breunig, R. A. Saroyan, R. Nusshardt, Q. C. Johnson, F. Busch, and S. D. Antolovich, Journal of Materials Research, 5(5), 1123-1129, 1990.
293. R. N. Yancey, and G. Y. Baaklini, Journal of Engineering for Gas Turbines and Power, Transactions of the ASME, 116(3), 635-639, 1994.
294. S. R. Stock, L. L. Dollar, G. B. Freeman, W. J. Ready, L. J. Turbini, J. C. Elliott, P. Anderson, and G. R. Davis, Proceedings of the Fall 1993 MRS Meeting, Nov 29-Dec 3 1993, Boston, MA, USA, Publ by Materials Research Society, Pittsburgh, PA, USA, p. 65-69, 1994.
295. L. L. Dollar, "Evaluation of nondestructive x-ray techniques for electronic packaging materials", Georgia Institute of Technology, MS Thesis, 1992.
296. C. F. Martin, C. Josserond, L. Salvo, J. J. Blandin, P. Cloetens, and E. Boller, Scripta Materialia, 42(4), 375-381, 2000.
297. A. Elmoutaouakkil, L. Salvo, E. Maire, and G. Peix, Advanced Engineering Materials, 4(10), 803-807, 2002.
298. P. Cloetens, W. Ludwig, D. Van Dyck, J. P. Guigay, M. Schlenker, and J. Baruchel, Proceedings of SPIE - The International Society for Optical Engineering; Proceedings of the 1999 Developments in X-Ray Tomography II, Jul 22-Jul 23 1999, 3772, 279-290, 1999.
299. P. Cloetens, W. Ludwig, J. Baruchel, D. Van Dyck, J. Van Landuyt, J. P. Guigay, and M. Schlenker, Applied Physics Letters, 75(19), 2912-2914, 1999.
300. W. Ludwig, and D. Bellet, Materials Science and Engineering A: Structural Materials: Properties, Microstructure and Processing, 281(1), 198-203, 2000.
References
419
301. W. Ludwig, J.-Y. Buffiere, S. Savelli, and P. Cloetens, Acta Materialia, 51(3), 585-598, 2003.
302. R. J.-P. Doyle, "The Effect of Welding Parameters, Laser Surface Melting Techniques and Shot Peening on the Corrosion Behaviour and Fatigue Properties of Friction Stir Welded AA2024-T351", University of Birmingham, Masters of Research Thesis, 2006.
303. D. R. Lide, CRC Handbook of Chemistry and Physics, CRC Press, Online Edition, 2000.
304. K. S. Ferrer, and R. G. Kelly, Corrosion, 57(2), 110-117, 2001. 305. T. Suter, and H. Bohni, Electrochimica Acta, 43(19-20), 2843-2849, 1998. 306. T. Suter, Y. Mueller, P. Schmutz, and O. von Trzebiatowski, Advanced
Engineering Materials, 7(5), 339-348, 2005. 307. M. M. Lohrengel, A. Moehring, and M. Pilaski, Electrochimica Acta, 47(1-2),
137-141, 2001. 308. M. M. Lohrengel, C. Rosenkranz, I. Kluppel, A. Moehring, H. Bettermann, B. V.
d. Bossche, and J. Deconinck, Electrochimica Acta, 49(17-18), 2863-2870, 2004. 309. N. Birbilis, B. N. Padgett, and R. G. Buchheit, Electrochimica Acta, 50(16-17),
3536-3544, 2005. 310. N. Mary, V. Vignal, R. Oltra, and L. Coudreuse, Journal of Materials Research,
19(12), 3688-3694, 2004. 311. C. Padovani, "Corrosion Protection of Friction Stir Welds in Aerospace
Aluminium Alloys", University of Birmingham, PhD Thesis, 2007. 312. J. F. Nye, Physical Properties of Crystals: Their Representation by Tensors and
Matrices, Oxford University Press, Oxford, 1985. 313. M. Jariyaboon, A. J. Davenport, R. Ambat, B. J. Connolly, S. W. Williams, and D.
A. Price, Corrosion Science, 49(2), 877-909, 2007. 314. L. A. Feldkamp, L. C. Davis, and J. W. Kress, Journal of the Optical Society of
America A (Optics and Image Science), 1(6), 612-619, 1984. 315. C. Blanc, B. Lavelle, and G. Mankowski, Materials Science Forum, 217-222(3),
1559-1564, 1996. 316. S. Hoerle, B. Malki, and B. Baroux, Journal of the Electrochemical Society,
153(12), B527-B532, 2006. 317. C. Blanc, A. Freulon, M.-C. Lafont, Y. Kihn, and G. Mankowski, Corrosion
Science, 48(11), 3838-3851, 2006. 318. K. R. Trethewey, and J. Chamberlain, Corrosion for Science and Engineering,
Addison Wesley Longman Limited, Essex, UK, 1998. 319. X. Liu, G. S. Frankel, B. Zoofan, and S. I. Rokhlin, Journal of the
Electrochemical Society, 153(2), 42-51, 2006. 320. W. Zhang, and G. S. Frankel, Journal of the Electrochemical Society, 149(11),
B510-B519, 2002. 321. C. Augustin, E. Andrieu, C. Blanc, G. Mankowski, and J. Delfosse, Journal of the
Electrochemical Society, 154(11), C637-C644, 2007. 322. R. H. Jones, M. A. Friesel, and R. Pathania, Corrosion (Houston), 47(2), 105-115,
1991. 323. T. Suter, E. G. Webb, H. Bohni, and R. C. Alkire, Chemistry and
Electrochemistry of Corrosion and Stress Corrosion Cracking, New Orleans, LA, p. 301-317, 2001.
324. T. K. Christman, Corrosion (Houston), 46(6), 450-453, 1990.
References
420
325. T. P. Hoar, D. C. Mears, and G. P. Rothwell, Corrosion Science, 5(4), 279-289, 1965.
326. G. T. Burstein, and S. P. Mattin, Critical Factors in Localized Corrosion II, Chicago, IL, USA, Electrochem. Soc, p. 1-14, 1996.
327. S. P. Mattin, and G. T. Burstein, Philosophical Magazine Letters, 76(5), 341-347, 1997.
328. T. P. Hoar, Corrosion Science, 7(6), 341-355, 1967.